JOURNAL OF EXPERIMENTAL ZOOLOGY 286:219–230 (2000) Morphology and Physiology of Auditory and Vibratory Ascending Interneurones in Bushcrickets BERND NEBELING* AG Neurobiologie, FB Biologie, Philipps Universität, 35033 Marburg, Germany ABSTRACT Auditory/vibratory interneurones of the bushcricket species Decticus albifrons and Decticus verrucivorus were studied with intracellular dye injection and electrophysiology. The morphologies of five physiologically characterised auditory/vibratory interneurones are shown in the brain, subesophageal and prothoracic ganglia. Based on their physiology, these five interneurones fall into three groups, the purely auditory or sound neurones: S-neurones, the purely vibratory Vneurones, and the bimodal vibrosensitive VS-neurones. The S1-neurones respond phasically to airborne sound whereas the S4-neurones exhibit a tonic spike pattern. Their somata are located in the prothoracic ganglion and they show an ascending axon with dendrites located in the prothoracic, subesophageal ganglia, and the brain. The VS3-neurone, responding to both auditory and vibratory stimuli in a tonic manner, has its axon traversing the brain, the suboesophageal ganglion and the prothoracic ganglion although with dendrites only in the brain. The V1- and V2neurones respond to vibratory stimulation of the fore- and midlegs with a tonic discharge pattern, and our data show that they receive inhibitory input suppressing their spontaneous activity. Their axon transverses the prothoracic ganglion, subesophageal ganglion and terminate in the brain with dendritic branching. Thus the auditory S-neurones have dendritic arborizations in all three ganglia (prothoracic, subesophageal, and brain) compared to the vibratory (V) and vibrosensitive (VS) neurones, which have dendrites almost only in the brain. The dendrites of the S-neurones are also more extensive than those of the V-, VS-neurones. V- and VS-neurones terminate more laterally in the brain. Due to an interspecific comparison of the identified auditory interneurones the S1-neurone is found to be homologous to the TN1 of crickets and other bushcrickets, and the S4-neurone also can be called AN2. J. Exp. Zool. 286:219–230, 2000. © 2000 Wiley-Liss, Inc. In bushcrickets, the tympanal and the subgenual organs function in the perception of airborne sound and substrate borne vibration respectively (Schumacher, ’73; Rössler, ’92). The auditory and vibratory receptor cells involved in this process are located in the proximal parts of the anterior tibiae and are known to converge synaptically to higher order auditory and/or vibratory neurones (Kalmring et al., ’79; Römer et al., ’88). Although in bushcrickets the number of receptor cells (Kalmring et al., ’93) and the number of ascending neurones has been shown to be species specific, receptor cells are always more numerous than interneurones. Therefore, sensory input must converge at the level of the dendritic region of a relevant interneurone. Ascending interneurones in the ventral nerve cord of bushcrickets have been subject to detailed morphological and physiological research (Römer et al., ’88; Shen, ’93). In the brain and subesophageal ganglion (SEG), however, interneurones © 2000 WILEY-LISS, INC. have only been described physiologically (Rheinlaender and Kalmring, ’73). Very poor evidence is available on the morphology of the auditory interneurones in the brain, and so far no research has been carried out to follow the projections of the arborizations of vibratory interneurones in the brain. By investigating both the physiology and morphology of ascending auditory and vibratory interneurones in the brain this study aimed to fill gaps in our knowledge of the organization of the auditory and vibratory pathways in Tettigoniids. MATERIALS AND METHODS Animals and preparation Experiments were performed on sexually mature male and female Decticus albifrons and Grant sponsor: Sonderforschungsbereich Ökophysiologie (305); Grant number: Teilprojekt A7. *Correspondence to: Bernd Nebeling, MZ für Augenheilkunde, Philipps Universität, Robert-Koch-Str. 4, 35033 Marburg, Germany. Received 7 April 1999; Accepted 17 June 1999 220 B. NEBELING Decticus verrucivorus obtained from their natural habitat in southern France and Bavaria. The animals were kept in the laboratory for up to four weeks under appropriate conditions (Rheinlaender and Kalmring, ’73). Prior to the experiments the animals were anaesthetised with CO2 and dorsally attached to a holder. All legs were fixed, placing the forelegs to the wire holder of a mini-shaker in a natural position. The animals were then dissected from the ventral surface, the gut and the mouthparts were removed, and the caudal part of the suboesophageal ganglion (SEG) together with the connectives were exposed. The experiments comply with the NIH “Principles of animal care” and current German laws. Electrophysiology and staining For intracellular recordings, microelectrodes were drawn from thick-walled borosilicate glass microcapillaries. They were filled with 1.5 M NiCl2 and had a resistance of about 20 MΩ. Alternatively, the tip of the microcapillaries was filled with 5% Lucifer-Yellow (Sigma), backed up with 0.2 M LiCl solution (80 MΩ). In the latter, for iontophoretic injection a constant hyperpolarizing current of 2–5 nA was set for up to 1 hr. The NiCl2-technique appears to be more suitable for recording from interneurones of very small diameter, whereas the method using Lucifer-Yellow is more advantageous for stainings over long distances. Therefore, in this study both techniques were applied. The indifferent electrode was inserted into the abdomen. All experiments were carried out in an anechoic chamber which is acoustically insulated against background noise. Standard airborne-sound and substrate-borne sound stimuli (duration 20 ms and 100 ms, rise/ fall time 1 ms, repetition rate 2/s) of different frequencies were used. The signals were generated by two acoustic stimulators (Burchard II). The acoustic stimuli were delivered ipsilaterally to the site of penetration in a distance of 41 cm from the penetration via a HF-loudspeaker (Audax, TW 8), perpendicular to the body axis of the animal. The vibration was passed to the fore- and midlegs by a minivibrator (Bruel + Kjaer, Typ 4810). The intensities of the sound signals were calibrated with a condenser microphone (Bruel + Kjaer 4135 or 4138) and a B + K measuring amplifier (2209) using its “peak hold” function. To ensure comparability, all signal intensities are given in dB peak SPL, which is for sine waves 3 dB above the respective RMS value. Sound mea- surements were obtained on the preparation site, with no animal present. The responses of each single cell were stored on magnetic tape (Racal 4DS) and later replayed, digitized (sampling frequency 200 or 400 kHz) and analyzed by custom made computer programs. The responses are presented as spike-trains and peri-stimulus-time-histograms (PSTHs). Threshold curve and response magnitude are given for each interneurone, and the neurones were classified physiologically according to Silver et al. (’80). Histology After the electrophysiological experiments, the brain, SEG, and first two thoracic ganglia were dissected. When NiCl2 had been employed, the nickel-ions were precipitated with rubeanic acid. After fixation, silver intensification, and dehydration, the ganglia were cleared in styrene (Bacon and Altman, ’79). The projection patterns of the interneurones were then reconstructed in whole mounts via camera lucida drawings. For Lucifer-Yellow staining the ganglion chain was fixed in phosphate-buffered 4% formalin for 10 min and afterwards dehydrated and transferred to 4% formalin in alcohol for at least 1 hr. The ganglia were rinsed in alcohol (100%) and stored in methyl salycilate. Whole mounts were photographed in a series of focal planes in the fluorescence microscope. In addition, for permanent visualization the ganglia were embedded in gelatin/albumin and serial thick sections (20 µm) were obtained with a vibratome. Using the PAPtechnique (Peroxidase-Anti-Peroxidase) the staining of the interneurones could be intensified (Anti-Lucifer, rabbit, Sigma; second antibody: Goat Anti Rabbit; third step: rabbit PAP, intensification with DiAmino-Benzidin; for more details see Nebeling, ’95). A frontal view of the morphology of each neurone was drawn by using both whole mounts and serial-sections which allowed tracing very fine branches. A sagittal view was reconstructed using only the serial sections and this allowed tracing only the main branches. RESULTS The nomenclature in this work follows the physiological classification of Kühne et al. (’80) and Silver et al. (’80) because vibrosensitive and bimodal neurones have so far not been characterised morphologically. AUDITORY NEURONES IN THE BRAIN OF BUSHCRICKETS As no sex specific differences were obvious, male and females were not treated separately. Physiologically data of almost all types of interneurones described in literature could be recorded in both species—no major differences were found. Additionally, some neurones of the same type stained in D. albifrons and in D. verrucivorus showed very similar arborization patterns. Therefore, the results of both species were not treated separately. Based on their physiological responses the interneurones were divided into three groups: S, VS and V, and each is described separately. S-neurones These interneurones are characterized by their exclusive response to airborne sound (Silver et al., ’80) and comprise five different subtypes which are distinguished by their specific spike patterns. Two of these subtypes were identified in this work. S-neurone, type S1 The S1-neurone has a large fiber diameter and has therefore already been studied before by several authors (e.g., Kalmring et al., ’79). Following a short habituation phase (not shown in the figures) the S1-neurones (n = 21) responded to airborne sound in a strictly phasic manner. The responses remained phasic even to stimuli of long duration (100 ms). The neurones were sensitive to a broad band of frequencies (Fig. 1A–C). In the most sensitive frequency band between 12 and 30 kHz the neurones responded to sound of less than 40 dB SPL intensity. The response latency was short (12 ± 0.5 ms SD) in keeping with the large diameter of these neurones and no spontaneous activity was detected (Fig. 1A, B). S1-neurones were injected with dye in the SEG from where the dye ascended to the terminations in the cephalic ganglia and descended backwards to the prothoracic ganglion, where it filled a contralaterally located somata (Fig. 1D). In the SEG, 8 to 10 collaterals form an extensive area of end-branches ipsilaterally to the axon. In the brain, S1-neurones terminate in the median parts of the ventro-lateral protocerebrum. In contrast to former investigations (Kalmring et al., ’79) an additional arborization area was stained in the median deutocerebrum. The spatial relation of the terminations to some major brain structures (mushroom body, central body, protocerebral bridge) of one representative S1-neurone is shown in both frontal and sagittal views (Fig. 1E, F) and show that terminations of this neurone do not enter these brain structures. 221 S-neurone, type S4 Neurones of the type S4 were recorded (n = 5) in the median part of the neck-connectives, closely lateral to the S1-neurones. The diameter of their axon is large but smaller than that of the S1. The physiology of S4-neurones is demonstrated (Fig. 2A–C). The most obvious difference to type S1neurones is in their tonic spike-pattern following stimuli of 20 ms and 100 ms duration. Almost no habituation was found. The latency of these neurones (19.5 ± 0.5 ms SD) was longer than the latency of the S1-type; also the spike repetition rate was higher. Like S1-neurones they responded to sound over the whole frequency range from 3 to 60 kHz with best frequencies from 10 to 25 kHz. The morphology of S4-neurones closely resembles that of S1-neurones. Thus the somata are located contralaterally to the axon in the anterior cortex of the TH1 (Fig. 2D) and a single axon ascends to the brain. Along the way the axon originates 5 to 7 (ipsilateral) collaterals in the SEG, oriented towards the midline similar to those in S1 although not as densely branched as those of S1. In the protocerebrum the axons of these S4-neurones form extensive terminations extending into the optic stalk (Fig. 2E, F). Some collaterals terminate in the median deutocerebrum building an additional, separate projection area. The terminations of the S4 neurones do not contact the mushroom body or the central body. VS-neurones These bimodal interneurones which show excitatory responses to both airborne sound and vibration were named VS-neurones by Silver et al. (’80) and five subtypes can be distinguished by their spike patterns. One of these types, the VS3neurone, is described here. The physiological data of VS3 neurones are shown in Figure 3A–C. They responded tonically to sound and (phasic)-tonically to vibratory stimulation. In their responses to airborne sound they resembled the S4 type. In contrast to S4-neurones, VS3-neurones showed spontaneous activity and exhibited a greater latency of 29 ± 2 ms (SD). Spontaneous-activity was suppressed for more than 200 ms after stimulation with stimuli of high efficiency (e.g., poststimulus suppression occurs to vibration of 200 Hz and high intensity). These neurones were sensitive to vibratory stimuli over the whole tested range from 100 to 2000 Hz, having a threshold acceleration of at least 0.03 m/s2. 222 B. NEBELING Fig. 1. Morphology and physiological data of an S1-neurone (TN1). (A) Five responses to repetitive stimulations with white noise (WN, 20 ms duration, 73 dB SPL). (B) Spike trains (upper part) and PSTH (2 ms bin width) to white noise (WN, 100 ms duration, 73 dB SPL). (C) Response magnitude to stimulation with airborne sound. (+) = tested without excitatory effect. (D) Position and morphology within the head and prothoracic ganglia, site of penetration (which is the same for all experiments) is indicated. Arrow indicates an additional arborization area, which was not stained in earlier investigations. (E, F) Reconstruction of the endbranches within the brain and relationship to prominent structures; (E) frontal view, (F) sagittal view. AUDITORY NEURONES IN THE BRAIN OF BUSHCRICKETS Fig. 2. Morphology and physiological data of an S4-neurone (AN2). Spike trains (upper part) and PSTH (2 ms bin width) to stimulation with (A) white noise (WN, 20 ms duration, 73 dB SPL); (B) white noise (WN, 100 ms duration, 73 dB SPL). (C) Response magnitude to stimulation with airborne sound. (+) = tested without excitatory effect. (D) Posi- 223 tion and morphology within the head and prothoracic ganglia. Arrow indicates an additional, separate projection area. (E, F) Reconstruction of the endbranches within the brain and relationship to prominent structures (E) frontal view, (F) sagittal view. 224 B. NEBELING Fig. 3. Morphology and physiological data of a VS3-neurone. Spike trains (upper part) and PSTH (2 ms bin width) to stimulation with (A) white noise (WN, 20 ms duration, 73 dB SPL); (B) vibration (200 Hz, 100 ms duration, 2.7 m/s2); arrows indicate the suppression of spontaneous activity. (C) Response magnitude to stimulation with airborne sound (up- per diagram) and vibration. (+) = tested without excitatory effect. (D) Position and morphology within the head and prothoracic ganglia. (E, F) Reconstruction of the endbranches within the brain and relationship to prominent structures; (E) frontal view, (F) sagittal view. AUDITORY NEURONES IN THE BRAIN OF BUSHCRICKETS Stainings of VS-neurones could be obtained in the brain and the thoracic ganglia in four animals. The reconstruction of a representative VS3neurone in the brain, SEG and the TH1 is demonstrated in Figure 3D and did not reveal a soma in any of those regions. Moreover, the axon did not reveal any collaterals in the SEG even though fine branches were stained in the brain. Thus the lack of collaterals in the SEG appears real. In the brain, the VS3-neurones terminate in the outermost layer of the ventro-lateral protocerebrum in the form of a loop with some branches running along the outer periphery of the ganglion. The terminations of the VS3-neurones do not contact any of the central brain structures shown (Fig. 3E, F). V-neurones These neurones respond to vibrational stimulation, but not to sound-stimuli at physiological intensities and may also be subdivided into five types (Silver et al., ’80). The type V1 neurone responded excitatorily to vibrational stimuli (n = 5). However, spontaneous activity was inhibited after sound signals of high intensity (Fig. 4A). Also vibration-stimuli inhibited the spontaneous activity. Spontaneous activity was suppressed before and after the excitatory vibration response. The latency up to the beginning of the inhibition (20 ± 2 ms SD) (measured up to the end of the last spontaneous spike) was shorter than that of the excitatory response to vibratory stimuli (52 ± 2 ms SD). The PSTHs show that the spontaneous activity returned after 150 ms (Fig. 4A) and 300 ms (Fig. 4B). These data demonstrate that the inhibition occurring before the excitatory response did not result from poststimulus suppression of the previous stimulus. The neurones did not habituate to repetitive stimuli. The responses of neurones of this type are excitatory from 50 to 500 Hz (Fig. 4C). Spontaneous activity was inhibited from 600 to 2000 Hz. In the SEG a few collaterals project towards the midline, but they do not cross it (Fig. 4D). In the brain, terminations are located along the outer periphery of the ventro-lateral protocerebrum. One branch describes a striking loop. A tiny collateral was stained in the deutocerebrum. Although some branches of these neurones lay in the same plane as the central body, none of the arborizations invade the brain structures shown (Fig. 4E, F). The V2 type neurone has physiological data (n = 2) very similar to that of the V1-type (Fig. 5A– 225 C). If vibratory stimuli were presented, the spiketrains of this spontaneous active neurones showed tonic discharge patterns. Spontaneous activity was suppressed before and after the excitatory vibration response. The responses of the V2-neurones had a latency of 33 (±2 SD) ms to the beginning of the inhibition, or 61 (±2 SD) ms up to the first spike (Fig. 5A). In contrast, vibration up to 2000 Hz lead to excitatory responses instead of inhibition (Fig. 5C). The soma of the V2-neurone could not be stained, it is likely to be located posterior to the prothoracic ganglion. In the SEG, four tiny ipsilateral collaterals were stained. Two of them are located in the more caudal and two in the more frontal part of the SEG. In the brain, the arborizations project into the ventro-lateral protocerebrum (Fig. 5D). In relation to the central body and mushroom body the terminations lie more laterally and caudally and do not overlap with these brain-structures (Fig. 5E, F). Comparative morphology The identified interneurones in bushcrickets show clear differences in the degree and location of projections from their axons, that fall into two categories. In the one category represented by the S-neurones there are projections in the SEG and in two locations (median deutocerebrum and ventro-lateral protocerebrum) in the brain (Fig. 6). In contrast, the other category represented by the VS- and V-neurones have either no or little projections in the SEG and projections in the brain to a single location (median deutocerebrum) where these projections are more laterally placed. The difference in the degree of the arborization among these two categories seen visually in the figures, is also reflected in both the surface area and volume measurements (Fig. 7). Thus the Sneurones have more extensive terminations than the VS- or V-neurones. DISCUSSION Although the morphology and physiology of auditory and vibratory interneurones is known in the prothoracic ganglion in a variety of bushcricket species (Rheinlaender and Kalmring, ’73; Kühne et al., ’80; Silver et al., ’80; Römer, ’87; Römer et al., ’88; Rheinlaender and Römer, ’86) their morphology in the SEG and the brain has been poorly investigated (Kalmring et al., ’79; Shen, ’93). This lack of knowledge is surprising as vibratory signals are important in acoustic behaviour (Latimer and Schatral, ’83; Stiedl and Kalmring, ’89). 226 B. NEBELING Fig. 4. Morphology and physiological data of a V1-neurone. Spike trains (upper part) and PSTH (2 ms bin width) to stimulation with (A) white noise (WN, 20 ms duration, 73 dB SPL); (B) vibration (200 Hz, 100 ms duration, 2.7 m/s2); arrows indicate the suppression of spontaneous-activity. (C) Response magnitude to stimulation with vibration. (+) = tested without excitatory effect. (D) Position and morphology within the head ganglia (TH1 was lost). Arrow indicates a branch describing a striking loop. (E, F) Reconstruction of the endbranches within the brain and relationship to prominent structures; (E) frontal view, (F) sagittal view. Arrow see (D). AUDITORY NEURONES IN THE BRAIN OF BUSHCRICKETS Fig. 5. Morphology and physiological data of a V2-neurone. Spike trains (upper part) and PSTH (2 ms bin width) to stimulation with (A) vibration (800 Hz, 100 ms duration, 2.7 m/s2); (B) vibration (100 Hz, 100 ms duration, 0.27 m/s2); arrows indicate the suppression of spontaneous activity. (C) 227 Response magnitude to stimulation with vibration. (+) = tested without excitatory effect. (D) Position and morphology within the head and prothoracic ganglia. (E, F) Reconstruction of the endbranches within the brain and relationship to prominent structures; (E) frontal view, (F) sagittal view. 228 B. NEBELING Cokl et al. (’77) suggested that information about airborne sound and substrate-borne vibration may converge from the receptor cells to the same auditory/vibratory interneurones. How this converged information may subdivide in the brain and SEG could be reflected in the morphology. S-neurones Fig. 6. Positions of two representative neurones, one of the auditory (S1 = TN1) and one of the vibratory subsystem (V1) put together into one shape of the brain. For a better comparison, the V1-neurone was mirror-imaged. Based on their physiology, morphology, and soma position the S1-neurone is found to be synonymous to the TN1 and the S4-neurone is synonymous to the AN2 in Tettigonia viridissima (Römer and Marquart, ’84; Römer et al., ’88). The names TN and AN were first given for a homologous t-shaped neurone (TN) and an ascending neurone (AN) by Wohlers and Huber (’82) in crickets. The present study brings together different nomenclatures which so far have been used to describe synonymous neurones either morphologically or physiologically. In the prothoracic ganglion staining of the AN2 in various cricket and bushcricket species reveal very similar morphological features (Wiese, ’81; Römer and Marquart, ’84; Boyan, ’84; Römer et al., ’88; Mason and Schildberger, ’93). On the contrary, the morphology of the AN2 in the brain of Decticus albifrons differs from that of crickets (Boyan and Williams, ’82; Schildberger, ’84). The termination field of ascending interneurones seems to be specific to the crickets, for ascending auditory neurones of Locusta also project to the lateral protocerebrum (Eichendorf and Kalmring, ’80; Boyan, ’83). In general, because of the similarities in the TH1 one may conclude that the dendritic region of interneurones could reflect “old” morphological solutions, whereas in the brain the termination fields may vary depending on the evolutionary necessities. The present study supports the hypothesis that optical and auditory information may converge in the optic stalk within the termination areas of AN2 (S4) neurones. This pathway has already been suggested by Horridge (’64) in Acridids. Similarly, Eichendorf and Kalmring (’80) have reported that in Acridids auditory B1 neurones project into the lateral horn, close to the optic lobes. VS3-neurones Fig. 7. Size of the terminations of the neurone-types in the brain. Areas are measured with the computer program Auto-CAD. Volumes are calculated by multiplication of the areas with the numbers and the thickness of the slides (20 µm) in which terminations appear; relative units to be multiplied by 102 (area/µm2), respectively, 103 (volume/µm3). The soma of a VS3-neurone was not stained in any of the histological preparations, therefore no homonymous morphological name can be given. Most likely the somata of these neurones are posterior to the prothoracic ganglion. Shen (’83) stained some giant interneurones with somata lo- AUDITORY NEURONES IN THE BRAIN OF BUSHCRICKETS cated in the terminal abdominal ganglion in Tettigonia cantans and one of these neurones is very similar in morphology to the VS3-neurone. Moreover, these giant interneurones respond to wind input from the cerci of the same frequencies used for vibratory stimulation in our work. As the cerci are seen to be homologous to legs, the giant interneurones based in the abdominal ganglion might be regarded as homologous to the VS3neurones. The VS3-interneurones may receive sensory input in the more caudal part of the CNS (e.g., the mesothoracic or metathoracic ganglion) because of the relatively long latency (29 ± 2 ms) and the fact that there were no dendrites stained in the prothoracic ganglion. In this case information on airborne sound might have its input via another descending interneurone to the VS3-neurone. The fact that spontaneous activity is inhibited by either sound or vibrational stimuli could be explained by the influence of inhibitory interneurones. Stumpner and Ronacher (’91) found an inhibitory influence of airborne sound stimuli to spontaneous auditory neurones in Acridids, and Schul (’97) has reported inhibitory effects to the AN1 of Tettigonia viridissima. V-neurones For both of the V-neurones, a soma could not be located, although the axon of V2 stained in the prothoracic ganglion. Most likely the somata of both V-neurones are located in the more caudal part of the CNS. In both neurones spontaneous activity is inhibited even before the excitatory response occurs. One explanation could be that these neurones receive input from inhibitory neurones with shorter latency than the excitatory input. Comparison of auditory and vibrosensitive neurones From the morphological characteristics revealed in this study it becomes obvious that the two auditory interneurones (S1, S4) differ from the vibrosensitive cells (VS, V). In the SEG, the S1 and S4-neurones (TN1 and AN2) build extensive terminations, whereas the VS3-, V1-, and V2neurones have none or only weak arborizations (Figs. 1–5). Consequently, the influence of the VSand V-neurones on the activity of higher order interneurones would be limited. On the other hand, the SEG is known to receive the nerves innervating the mouthparts in its mandibular, maxillar and labiate neuromeres (Tyrer and Gregory, ’82). Also in Acridids and crickets, numer- 229 ous motoneurones innervating the neck-muscles are located in the SEG (Honegger et al., ’84). Since these animals protect themselves from predators by aggressive biting, there is likely to be a correlation between the behavioural element of defence and the activation of the mouthparts. However, whether that the S-neurones are involved in these behaviour is still to be investigated, although earlier investigations found the S1-neurone being related with the fight- or flight-reaction (McKay, ’70; Kalmring et al., ’79). The S-neurones build more extensive terminations than the VS and V-neurones do in the brain of bushcrickets (Fig. 7). The bigger spatial representation of the auditory interneurones may represent larger numbers of synaptic contacts on higher order interneurones, and this may compensate for the smaller number of auditory receptor cells. Moreover, both S-neurones build a second arborization area in the median deutocerebrum, which is missing in all V- or VS-neurones. The latter neurones also terminate more laterally in the brain than the S-neurones. The collective differences between the S-neurones and the VS-, Vneurones dealing respectively with auditory and vibratory stimuli suggest that these two sensory modalities are separated by different locations of the neuronal terminations in the brain and SEG. ACKNOWLEDGMENTS The author thanks Prof. Dr. K. Kalmring for supporting this work and Dr. T. Sickmann for critically reading the concept. C. Meyer improved the English text. I am also grateful to Drs. W. Rössler, M. Jatho, and J. Schul for the friendly gift of the computer software for analytical purposes. The author thanks Prof. Dr. U. Homberg and M. Müller for assistance with the antibody-procedure, and Mrs. S. Völk and H. Brandtner for technical help. Dr. R. Lakes-Harlan and Prof. Dr. Schildberger introduced me to intracellular recordings. LITERATURE CITED Bacon JP, Altman JS. 1979. A silver intensification method for cobaltfilled neurones in wholemount preparations. Brain Res 138:359–363. Boyan GS. 1983. Postembryonic development in the auditory system of the locust. J Comp Physiol A 151:499–513. Boyan GS. 1984. Neural mechanisms of auditory information processing by identified interneurons on orthoptera. J Insect Physiol 30:27–41. Boyan GS, Williams JLD. 1982. Auditory neurones in the brain of the cricket Gryllus bimaculatus (De Geer): ascending interneurones. J Insect Physiol 28:493–501. Cokl A, Kalmring K, Wittig H. 1977. The responses of auditory ventral-cord neurons of Locusta migratoria to vibration stimuli. J Comp Physiol 120:161–172. 230 B. NEBELING Eichendorf A, Kalmring K. 1980. Projections of auditory ventral-cord-neurons in the supraesophageal ganglion of Locusta migratoria. Zoomorphologie 94:133–149. Honegger HW, Altmann JS, Klein J, Müller-Tautz R, Pollerberg E. 1984. A comparative study of neck muscle motor neurons in a cricket and a locust. J Comp Neurol 230:517–535. Horridge GA. 1964. Multimodal interneurones of locust optic lobe. Nature 204:499–500. Kalmring K, Rehbein HG, Kühne R. 1979. An auditory giant neuron in the ventral cord of Decticus verrucivorus (Tettigoniidae). J Comp Physiol 132:225–234. Kalmring K, Rössler W, Ebendt R, Ahi J, Lakes R. 1993. The auditory receptor organs in the forelegs of bushcrickets: physiology, receptor cell arrangement, and the morphology of the tympanal and intermediate organs of three closely related species. Zool J Physiol 97:75–94. Kühne R, Lewis B, Kalmring K. 1980. The responses of ventral cord neurons of Decticus verrucivorus (L) to sound and vibration stimuli. Behav Proc 5:55–74. Latimer W, Schatral A. 1983. The acoustic behaviour of the bushcricket Tettigonia cantans: I. Behaviour responses to sound and vibration. Behav Proc 8:113–124. Mason AC, Schildberger K. 1993. Auditory interneurons in Cyphoderris monstrosa (Orthoptera: Haglidae). J Comp Physiol A 171:749–757. McKay JM. 1970. Central control of an insect sensory neuron. J Exp Biol 53:137–145. Nebeling B. 1995. Darstellung der Morphologie von physiologisch charakterisierten auditorischen Interneuronen im Kopfbereich (Ober- und Unterschlundganglion) der Beißschreckenarten Decticus albifrons und Decticus verrucivorus (Ensifera, Orthoptera). Göttingen: Cuvillier Verlag. 165 p. Rheinlaender J, Kalmring K. 1973. Die afferente Hörbahn im Bereich des Zentralnervensystems von Decticus verrucivorus (Tettigoniidae). J Comp Physiol 85:361–410. Rheinlaender J, Römer H. 1986. Insect hearing in the field: I. The use of identified nerve cells as “biological microphones.’’ J Comp Physiol A 158:647–651. Römer H. 1987. Representation of auditory distance within a central neuropil of the bushcricket Mygalopsis marki. J Comp Physiol A 161:33–42. Römer H, Marquart V. 1984. Morphology and physiology of auditory interneurons in the metathoracic ganglion of the locust. J Comp Physiol A 155:249–262. Römer H, Marquart V, Hardt M. 1988. Organization of a sensory neuropile in the auditory pathway of two groups of orthoptera. J Comp Neurol 275:201–215. Rössler W. 1992. Functional morphology and development of tibial organs in the legs, I, II and III of the bushcricket Ephippiger ephippiger (Insecta, Ensifera). Zoomorph 112: 181–188. Schildberger K. 1984. Temporal selectivity of identified auditory neurons in the cricket brain. J Comp Physiol A 155:171–185. Schul J. 1997. Neuronal basis of phonotactic behaviour in Tettigonia viridissima: processing of behaviourally relevant signals by auditory afferents and thoracic interneurons. J Comp Physiol A 180:573–583. Schumacher R. 1973. Beitrag zur Kenntnis des tibialen Tympanalorgans von Tettigonia viridissima L. (Orthoptera: Tettigoniidae). Mikroskopie 29:8–19. Shen JX. 1983. The cercus-to-giant interneuron system in the bushcricket Tettigonia cantans: morphology and response to low frequency sound. J Comp Physiol A 151:449–459. Shen JX. 1993. Morphology and physiology of auditory interneurons of the bushcricket Gampsocleis gratiosa. Japanese J Physiol 43(suppl 1):239–246. Silver S, Kalmring K, Kühne R. 1980. The response of central acoustic and vibratory interneurones in bushcrickets and locusts to ultrasonic stimulation. Physiol Entomol 5:427–435. Stiedl O, Kalmring K. 1989. The importance of song and vibratory signals in the behaviour of the bushcricket Ephippiger ephippiger Fiebig (Orthoptera, Tettinoniidae): taxis by females. Oecologica 80:142–144. Stumpner A, Ronacher B. 1991. Auditory interneurones in the metathoracic ganglion of the grasshopper Chorthippus biguttulus: I. Morphological and physiological characterization. J Exp Biol 158:391–410. Tyrer MN, Gregory GE. 1982. A guide to the neuroanatomy of locust suboesophageal and thoracic ganglia. Philos Trans R Soc Lond 297:91–123. Wiese K. 1981. Influence of vibration on cricket hearing: interaction of low frequency vibration and acoustic stimuli in the omega neuron (G. bimaculatus). J Comp Physiol A 143:135–142. Wohlers DW, Huber F. 1982. Processing of sound signals by six types of neurons in the prothoracic ganglion of the cricket Gryllus campestris L. J Comp Physiol A 146:161–173.