Auditory Neuropil Research Articles

Functional organization of the arctiid moth tymbal (insecta, lepidoptera).

The exoskeletal morphology, muscular organization, and innervation patterns of the tymbals of seven sound-producing species of tiger moths (Arctiidae) were compared with the undifferentiated episterna of two silent species. At least three muscles are involved in sound production: the tymbal muscle, pv2, and the accessory muscles, pvl and/or pv6. All of the tymbal muscles are innervated by the IIIN2a branch of the metathoracic leg nerve, which contains two axons larger than the others. Backfills of the tymbal branch of the IIIN2a reveal a medial sensory neuropil and a population of five ipsilateral motor neurons whose somata are clustered into three groups along the anterior edge of the metathoracic ganglion. The dendritic arborizations of the motor neurons extend to the ganglionic midline but are separate from one part of the auditory neuropil observed in other noctuoids. The study concludes that the arctiid tymbal reveals only minor modifications (e.g., cuticle thinning) of the episterna of silent moths and represents a primitive form of the tymbal compared to those of the Cicadidae.

Tympana, auditory thresholds, and projection areas of tympanal nerves in singing and silent grasshoppers (Insecta, Acridoidea)

The auditory systems of several species of singing and acoustically communicating grasshoppers, as well as of silent grasshoppers, were compared with respect to the external structure of the tympana, thresholds of the tympanal nerve response and projection areas of tympanal nerves within the metathoracic part of the ventral nerve cord. Extracellular recordings from the tympanal nerves, using suction electrodes, revealed that singing and silent grasshoppers hear within the frequency range tested, from 2 to 40 kHz. However, differences in sensitivity were observed in those silent species with tympana of modified structure. Cobalt-backfills of the tympanal nerves revealed a clearly discernible auditory neuropil in the anterior ring tract of the metathoracic ganglion in all animals. A comparison of the volumes of neuropilar areas calculated from serial sections of the entire ganglion showed a gradation: the volumes were biggest in singing species, slightly smaller in silent species with a well-developed tympanum, and smallest in the species with modified tympana. These findings support several authors who suggested that auditory organs evolved earlier than acoustic communication.

Effect of auditory deafferentation on the synaptic connectivity of a pair of identified interneurons in adult field crickets.

In adult crickets, Teleogryllus oceanicus, unilateral auditory deafferentation causes the medial dendrites of an afferent-deprived, identified auditory interneuron (Int-1) in the prothoracic ganglion to sprout and form new functional connections in the contralateral auditory neuropil. The establishment of these new functional connections by the deafferented Int-1, however, does not appear to affect the physiological responses of Int-1's homolog on the intact side of the prothoracic ganglion which also innervates this auditory neuropil. Thus it appears that the sprouting dendrites of the deafferented Int-1 are not functionally competing with those of the intact Int-1 for synaptic connections in the remaining auditory neuropil following unilateral deafferentation in adult crickets. Moreover, we demonstrate that auditory function is restored to the afferent-deprived Int-1 within 4-6 days following deafferentation, when few branches of Int-1's medial dendrites can be seen to have sprouted. The strength of the physiological responses and extent of dendritic sprouting in the deafferented Int-1 progressively increase with time following deafferentation. By 28 days following deafferentation, most of the normal physiological responses of Int-1 to auditory stimuli have been restored in the deafferented Int-1, and the medial dendrites of the deafferented Int-1 have clearly sprouted and grown across into the contralateral auditory afferent field. The strength of the physiological responses of the deafferented Int-1 to auditory stimuli and extent of dendritic sprouting in the deafferented Int-1 are greater in crickets deafferented as juveniles than as adults. Thus, neuronal plasticity persists in Int-1 following sensory deprivation from the earliest juvenile stages through adulthood.

The midline metathoracic ear of the praying mantis, Mantis religiosa.

The praying mantis, Mantis religiosa, is unique in possessing a single, tympanal auditory organ located in the ventral midline of its body between the metathoracic coxae. The ear is in a deep groove and consists of two tympana facing each other and backed by large air sacs. Neural transduction takes place in a structure at the anterior end of the groove. This tympanal organ contains 32 chordotonal sensilla organized into three groups, two of which are 180 degrees out of line with the one attaching directly to the tympanum. Innervation is provided by Nerve root 7 from the metathoracic ganglion. Cobalt backfills show that the auditory neuropile is a series of finger-like projections terminating ipsilaterally near the midline, primarily near DC III and SMC. The auditory neuropile thus differs from the pattern common to all other insects previously studied.

Regeneration of normal afferent input does not eliminate aberrant synaptic connections of an identified auditory interneuron in the cricket, Teleogryllus oceanicus.

In the cricket, Teleogryllus oceanicus, the dendritic arborizations of an identified auditory interneuron (Int-1) are normally restricted to the ipsilateral auditory neuropile; unilateral deafferentation causes the medial portion of the dendritic field to sprout across the midline and make functional connections with the contralateral auditory neuropile (Hoy et al., '78: Soc. Neurosci. Abstr. 4:115, '85: Proc. Natl. Acad. Sci. USA 82:7772-7786; Hoy and Moiseff, '79: Soc. Neurosci. Abstr. 5:163). We have found that regeneration of the auditory afferents also results in an aberrant pattern of innervation of Int-1. Crickets were unilaterally deafferented during postembryonic development by crushing or cutting the auditory nerve. Regeneration of afferent-to-Int-1 connections was tested behaviorally. Of 86 nerve-crushed crickets tested as adults in the behavioral assay, 66% showed functional regeneration of the afferents. Similar results were obtained from the nerve-cut group. However, morphological investigations demonstrated that most of the regenerates still retained the aberrant contralateral dendritic projection. Electrophysiological recordings from these Int-1s showed that not only are some of them driven by their regenerated auditory afferents (the normal pathway) but that they retain their excitability via their contralateral dendrites (the aberrant pathway). This demonstrates that reinnervation of Int-1 by its normal afferent pool neither causes retraction nor prevents the formation of connections made with foreign, contralateral afferents. When the contralateral afferent pool was removed after Int-1 had sprouted, the sprouts remained present, but preliminary results suggest that if the contralateral afferents are removed before Int-1 is deafferented, sprouts are not formed. The results are discussed in relation to the roles of competition and conservation of membrane area in controlling synapse formation.

Morphological and physiological changes in central auditory neurons following unilateral foreleg amputation in larval crickets

1. In view of the surprising recent demonstration that developmentally one-eared female crickets can track sound sources (Huber et al. 1984), we have looked for correlates in the morphological and physiological properties of auditory interneurons of these animals. One foreleg was amputated in the 3rd/4th or penultimate (8/9th) larval instar; in both cases the leg regenerated without developing a functional ear. In the adult stage, these animals were studied first for their phonotactic behavior and then by intracellular recording and staining; three types of auditory interneurons in the prothoracic ganglion were identified: the omega neuron ON1, and the ascending neurons AN1 and AN2. 2. Of these three neuron types, those that normally receive excitatory input from the side now deafferented send dendrites across the midline of the ganglion, along specific pathways, to end in the auditory neuropil of the intact side (Figs. 1–4). 3. The new connections are functional, as shown by the responses of the neurons to synthesized calling songs presented to the remaining ear. With respect to the copying of chirp structure, threshold curves and intensity characteristics, these neurons respond like cells in intact animals that are presented with the same stimulus on the side ipsilateral to the main input region of the neurons (Figs. 2–4). The implication is that in animals with one ear missing, functional pathways within the central nervous system are reorganized, resulting in better orientation of one-eared animals.

Dendritic sprouting and compensatory synaptogenesis in an identified interneuron follow auditory deprivation in a cricket.

We examined the effect of chronic afferent deprivation on an identified interneuron (Int-1) in the auditory system of the Australian field cricket Teleogryllus oceanicus. In normal intact crickets, the auditory afferents from each ear terminate ipsilaterally onto a single Int-1. Each bilaterally paired Int-1 is excited by ultrasound stimulation of its ipsilateral ear but not by the contralateral ear. Unilateral removal of an ear early in postembryonic development deprives the developing Int-1 of ipsilateral auditory innervation. Consequently, the ipsilateral dendrites of the deprived interneuron sprout, grow aberrantly across the ganglionic midline, and terminate specifically in the intact auditory neuropile of the contralateral (unlesioned) side, where they form functional synapses with the contralateral afferents. This unusual compensatory dendritic sprouting restores auditory function to the neuron. Thus, it is demonstrated that the dendritic shape of an identified Int, as well as its synaptic connectivity, is altered as a consequence of chronic sensory deprivation.

Morphology and physiology of auditory interneurons in the metathoracic ganglion of the locust

1. Auditory interneurons in the metathoracic ganglion of the locust were characterized by their intracellularly recorded responses to sound stimuli, and by their morphology as revealed in sectioned preparations of single neurons injected with Lucifer Yellow. Accordingly, nine interneurons were identified as either local, bisegmental, ascending or T-neurons (Figs. 2–6). 2. All cells possess arborizations within one or both sides of a prominent area of fine neuropil (frontal auditory neuropil), but the degree of overlap with the endings of tympanic afferents is different. For 2 ascending units (AN2, AN3) there is no overlap at all, indicating that they are interneurons of higher order. 3. Two neurons, SN1 and TN1, were found with only excitatory input to low frequency sound, and their tuning reflected that one of low frequency receptor fibres. In contrast, the response of all other neurons to low frequency stimuli was more complex, consisting of excitatory and inhibitory synaptic potentials. The neuronal circuitry which might underlie such physiological behaviour is discussed. 4. Recordings made from different parts of individual neurons enabled the sites of synaptic input and output to be localized. The postsynaptic dendrites of the interneurons appear to correspond to smooth endings in contrast to the beaded appearance of the presumptive presynaptic terminals (Fig. 8). 5. The results include descriptions of auditory ‘sister-neurons’ (Fig. 6) which have a similar morphology and of ‘twin-neurons’ (Fig. 7), which, using our present physiological and morphological criteria, cannot reliably be distinguished one from another. The problem that arises from such doubling of cells for the concept of identified units is discussed.

Tonotopic organization of the auditory neuropile in the bushcricket Tettigonia viridissima

Invertebrate central neural structures, especially sensory or motor neuropiles, have proved rather intractable to conventional physiological analysis as they are considered to have a diffuse or unordered structure. This is also true for the auditory neuropiles in the ventral nerve cord of orthopterans, where receptor fibres synapse to interneurones. In the bushcricket, Tettigonia viridissima, this neuropile constitutes the terminations of about 40 receptor fibres, which derive from chordotonal receptor cells in the tibia of the foreleg. Zhantiev1 and Rheinlaender2 have shown that each receptor cell is tuned to a ‘best frequency’. Oldfield3 has further shown that the sensilla are organized tonotopically within the organ, the proximal and distal receptors being tuned to low and high frequencies respectively. Besides the finding that auditory receptor axons form a diffuse neuropile on each side of the prothoracic ganglion4,5, there are no data on the projection of single receptor fibres within the neuropile. The results presented here show that the auditory neuropile of T. viridissima is tonotopically organized, extending to within the central nervous system the tonotopic organization of the peripheral sense organ.

Postembryonic development in the auditory system of the locust

1. The postembryonic development of three identified ascending interneurones in the auditory system ofLocusta migratoria was studied using intracellular recording and staining techniques. 2. The terminal arborisations in the brain of G, B and C neurones were identified in the adult, 5th and 4th instars, and for the G neurone in 3rd and 2nd instars as well (Figs. 2, 3). Further, the development of the soma and dendrites of the G neurone in the mesothoracic ganglion was followed from 2nd instar to adulthood (Fig. 4). 3. The adult G, B and C neurones all terminate in the lower lateral lobe of the protocerebrum, and G and B neurones share a further projection area ventrally in the anterior protocerebrum (Fig. 2). The lower lateral protocerebrum may represent an auditory neuropil in the brain. 4. The larval G, B and C neurones possess the general adult morphology throughout postembryonic development and neural growth is allometric. Only at the second instar level may some reduction in density of arborisation in brain and mesothoracic ganglion occur (Figs. 3, 4). 5. The physiological responses of the identified adult and larval neurones are very similar. Expressed firstly as constant intensity curves (frequency response spectra) (Figs. 5, 6), the shapes of the adult spectra are seen reflected in larval G, B and C neurones throughout development. With each step back in development there is an accompanying loss in sensitivity which affects higher more than lower frequencies and is not always linearly related to age (Fig. 8). The thresholds obtained from second instar animals suggest that newly hatched (first instar) locusts are probably deaf to airborne sound. 6. The intensity response and latency of response functions (Fig. 9 A) of the adult neurones are repeated in each larval instar studied; similarly the ipsilateral-contralateral difference characteristic of the B neurone is found intact in the 4th instar, although at a higher absolute intensity (Fig. 9B). These results suggest that the dynamic properties and synaptology described elsewhere for the adult neurones are already formed early in development. A role for these interneurones in the escape behaviour of the larva is considered.

Auditory neurones in the brain of the cricket Gryllus bimaculatus (De Geer): Ascending interneurones

Two types of auditory interneurone which ascend from the prothoracic ganglion to the brain in the cricket Gryllus bimaculatus (De Geer) are described. Intracellular recordings were made from the axons of the neurones in the brain under closed-field stimulus conditions and the recorded cells then stained with either cobalt or Lucifer Yellow. Both neurone types—the Plurisegmental ascending low frequency neurone 1 (PALF1), and the Plurisegmental ascending high frequency neurone 1 (PAHF1)—show response characteristics which would make them well suited to encoding the conspecific calling and courtship songs respectively. Further, the projection areas of both neurone types in the brain overlap those of previously identified intraganglionic interneurones, particularly in the anterior-ventral protocerebrum, and it is suggested that an auditory neuropile may exist in this region.

Auditory interneurones in the prothoracic ganglion of the cricket,Gryllus bimaculatus

1. A high-frequency auditory neurone (HF1AN) was identified morphologically and physiologically in the prothoracic ganglion of the cricket,Gryllus bimaculatus DeGeer. 2. The main inputs of the HF1AN are from the lateral and medial parts of the auditory neuropile ipsilateral to the axon and to a lesser extent, the medial part of the contralateral auditory neuropile. The single axon ascends to the brain and the soma is located either near the base of the anterior connective or close to the base of nerve N1 contralateral to the axon (Fig. 1). 3. The HF1AN is sensitive to sound stimulation in the whole range of tested (from 2 to 100 kHz) frequencies. The neurone is not tuned to a particular frequency and has greatest sensitivity in the range 13–25 kHz (Fig. 2). At frequencies below 10–12 kHz the neurone receives additional inhibitory inputs which greatly modify its response pattern from phasic-tonic with prolonged discharge to phasic with an inhibitory pause (Figs. 3–4). The inhibitory effect is intensity dependent and its frequency range varies between animals. In the range of frequencies where the inhibitory effect is pronounced, sound stimulation evokes suppression of activity during the sound pulse and increase of activity between the pulses. On the contrary, at high frequencies the cell gives a burst of spikes to each sound pulse and background activity (if present) is suppressed after each burst for at least 1–2 s (Figs. 5, 6). 4. The HF1AN responds in a phasic-tonic manner to pure high-frequency tones of long duration (1,300 ms) throughout the whole intensity range (Fig. 7). 5. The HF1AN responds strongly to high-frequency ticks of the courtship song of conspecific males and to the location sounds of sympatric bats (Fig. 8). 6. Both anatomical and physiological properties indicate that the HF1AN may belong to the neuronal circuits controlling evasive behavior and/or arousal of the cricket.

Auditory nerve and interneurone responses to natural sounds in several species of cicadas

ABSTRACT. The calling and courtship songs of 17‐year cicadas and of Say's cicadas differ both in the sound frequency spectrum and in temporal pattern. Multiunit recordings with hook electrodes from the whole auditory nerve show that the hearing organs are especially sensitive to transient stimuli occurring in natural sounds. Artificially produced clicks elicit bursts of spikes synchronized among various primary sensory fibres. These fibres respond to natural calling and courtship songs with a specificity dependent on carrier frequency, rhythm and transient content of the sound, following sound pulses (i.e. tymbal actions) up to repetition rates of 200 Hz. An ascending, plurisegmental interneurone was characterized by intracellular recording and simultaneously stained with cobalt. Its main arborization spatially overlaps the anterior part of the sensory auditory neuropile, and the axon was traced as far as the prothoracic ganglion. Direct input from primary auditory fibres was suggested by latency measurements. Intracellular recordings from such neurons in different species show distinct auditory input, with phasic‐tonic spike responses to tones. In general, the interneurone response is more species‐specific to calling than to courtship songs, and the preferential response to the conspecific calling song is based primarily upon sound frequency content.

Primary auditory neurons in crickets: Physiology and central projections

1. The tibial tympanal organs ofGryllus campestris L. andGryllus bimaculatus DeGeer were stimulated with sound signals simulating the natural calling song, produced by sinusoidal amplitude envelope, using carrier frequencies (CF) in the range of 2–20 kHz at various intensities. 2. Glass microelectrodes (filled with 3 M potassium acetate, or with 1 M cobalt nitrate for marking the fibers) were used to record extracellularly or quasi intracellularly the spike activity of single sensory fibers in the leg nerve (Fig. 1) responding to acoustic stimuli and to stain their central projections within the prothoracic ganglion (Figs. 6–9). 3. The temporal pattern of the sound stimulus —syllable and verse structure — is reflected in the discharge pattern of the fibers studied (Figs. 1, 6, 7). 4. Five types of fibers tuned to specific CFs were found (Fig. 2). These had peak sensitivities (a) at 4–5 kHz (CF of the species calling song), (b) at both 4–5 and 10–12 kHz, (c) below 2 kHZ, (d) at 12 kHz, and (e) at 17 kHz (CF of the species courtship song). 5. The tuning is also evident in the responsiveness (spikes/syllable) to stimuli of different frequencies at suprathreshold intensities (Fig. 3). 6. In the range of their best frequencies (BF) the fibers tuned to 4–5 and to 10–12 kHz are similar in intensity characteristics and in relationship of latencies to sound intensity, with a linear relationship between responsiveness and latency (Fig. 4). Within a single calling-song verse, the spike response decreases, and the latency increases, from the first to the fourth syllable (Fig. 5). Even at the highest intensities the discharge rate nerver exceeds 350 Hz. 7. The central arborization and projection areas of all the marked fibers are restricted to the ipsilateral half of the prothoracic ganglion. Assuming that fibers marked include those recorded physiologically, it appears that those tuned to 4–5 kHz terminate exclusively in the “crescent-shaped” neuropile, localized and termed the “auditory neuropile” (Figs. 6, 7). Fibers with a peak sensitivity below 2 kHz differ distinctly from the 4–5 kHz group in that they have additional arborizations outside the auditory neuropile (Fig. 8). As yet no unequivocal data are available regarding the spatial arrangement of the fibers with two sensitivity peaks (both 4–5 and 10–12 kHz) and those tuned to 12 kHz or 17 kHz (Fig. 9).

Auditory neurones in the brain of the cricketGryllus bimaculatus (De Geer)

1. Auditory neurones in the brain (supraoesophageal ganglion) of the cricketGryllus bimaculatus (De Geer) were intracellularly recorded and stained with Cobalt or Lucifer. 2. The cell bodies and arborisations of all recorded neurones are contained entirely within the supraoesophageal ganglion. Further, the cells all have arborisations which overlap extensively in the ventral-anterior protocerebrum, but there is no direct contact with the mushroom bodies. 3. The auditory neurones described fall into three morphological types. A. The first morphological type — theUnisegmentalauditorybrainneurone (UABN) — is restricted entirely to the protocerebrum and has no identifiable axon or terminal arborisations. Its Best Frequency (BF) is 16 kHz, near the conspecific courtship song frequency, and it is inhibited by low frequency sounds at high intensities. The intensity-response curve at 16 kHz has a greater dynamic range than those at higher or lower frequencies. B. The second morphological type — thePlurisegmentalauditorybrainneurone 1 (PABN1) — runs ventro-dorsally from proto- to deutocerebrum. The PABN1 has a bimodal tuning curve with a BF at 16 kHz and a secondary peak at 10 kHz. The intensity-response curve showing the greatest dynamic range occurs at 16 kHz. C. The third morphological type — thePlurisegmentalauditorybrainneurone 2 (PABN2) — sends branches from proto- to deutocerebrum, collaterals to its protocerebral and deutocerebral arborisations, and branches into the contralateral half of the brain. The PABN2 has a broader tuning curve than the other cells but is still most sensitive around 16 kHz. 4. All the cells exhibit spiking responses to sound, with tonic or phasic/tonic bursts which may exceed the duration of shorter stimulus tones. Sensitivities of the brain cells at their BF's are similar to those of receptors and ascending interneurones in this species. 5. All the brain cells recorded here can follow the pulse pattern of a synthetic calling song, adapting throughout a chirp, but only when chirps are presented at the BF. The initial response pattern is maintained over successive chirp presentations. This frequency-dependent encoding of the temporal properties of a stimulus is independent of sound intensity and, therefore, of auditory threshold. The underlying mechanism is probably different for each cell type. 6. The brain neurones have response properties and areas of arborisation which are similar to those of a previously identified ascending auditory interneurone. This ascending interneurone is thought to play a role in encoding the conspecific courtship song, with which these brain neurones may also be involved. The ventral-anterior protocerebrum may therefore represent a localised high-frequency auditory neuropile.

Central projections of tibial sensory fibers within the three thoracic ganglia of crickets (Gryllus campestris L.,Gvyllus bimaculatus DeGeer)

Cobalt sulfide (CoS) staining, via axonal filling, was used to reveal the central projections of fibers of the tympanal nerves in the cricket prothoracic ganglion, as well as those of serially homonymous nerves in the mesothoracic and metathoracic ganglia. The main results are: 1. The branch of the anterior leg nerve termed the tympanal nerve (TB in Fig. 2) comprises a few motor fibers and many sensory axons, as do serially homonymous nerves in the two other thoracic segments. The sensory axons are those of sense cells in the tympanal organ (or the serially homonymous parts of the meso and metathoracic segments), the subgenual organ, the campaniform sensilla, and the tibial sensory hairs (Eibl, 1978). 2. On the basis of their branching and projections in the three thoracic ganglia, the sensory fibers of nerve TB can be classified into five types (Figs. 3–7). With respect to function, those of Type 5 can be regarded as auditory fibers (Fig. 10). They project only to the crescent-shaped region (Fig. 8), which represents the auditory neuropile in the strict sense, and are tuned to the carrier frequency of the calling song (Fig. 10B). It is characteristic of Types 1 to 4 that the fibers branch immediately after entering the ganglion (Figs. 3–6); typically, these branches run dorsally and may overlap with the lateral projections from the motor neuropile. In addition, the fibers of Types 2 to 4 project into the rostral and caudal margins of the auditory crescent (Figs. 4–6). 3. The similarity among the three thoracic ganglia with respect to sensory fiber types and projections (Fig. 11) suggests that serially homonymous sensory structures found in each of the leg-bearing segments resemble one another in their sensory projection areas and thus in their connectivity with other neural subsystems within the three thoracic ganglia. That ‘typical auditory nerve fibers’ occur in the meso and metathoracic segments, projecting exclusively to the crescent (Figs. 7 and 11), is anatomical evidence that auditory function need not be restricted to the first thoracic segment. 4. The stained leg motoneurons (Fig. 12), with axons that innervate the femur muscles, are distinguished by a dendritic field that must be assigned to the ventral part of the so-called sensorimotor integration zone (Altman, 1976). These fields are of lesser extent than, for example, those of the flight and singing motoneurons of meso and metathorax. 5. All the sensory and motor elements marked with CoS are bilateral and arranged with approximately mirror-image symmetry. Their central projections are restricted entirely to the ipsilateral half of the ganglion (Fig. 13).

The afferent auditory pathway in the ventral cord oflocusta migratoria (Acrididae)

The afferent auditory pathway in the region of the ventral cord ofLocusta migratoria has been studied electrophysiologically. Various recording, stimulating and surgical techniques have been used to infer the circuitry underlying the response patterns of the ventral-cord neurons ascending to the supraesophageal ganglion. The stimuli used were artificial sounds (white noise and pure tones varying in frequency, intensity, duration, source direction and repetition rate). The responses of the auditory ventralcord neurons over the range of frequencies and intensities are described. Presentation of two sound stimuli occurring simultaneously or displaced in time permits the demonstration of inhibitory and subthreshold facilitatory influences which, with extracellular recording, cannot be observed directlyvia changes in membrane potential (Figs. 3 and 4). On the basis of comparisons of responses before and after inactivation of a tympanal organ additional inferences are made concerning the relative contributions of the two tympanal organs to the formation of the response-patterns (Figs. 6, 11, 16, 21, 25). The geometry, location and course of certain auditory ventral-cord neurons have been revealed by a combined recording and staining technique (Fig. 7); transection of connectives has made it possible to determine the origin of certain response-patterns and whether these change at various stations of the ventral cord as a result of corresponding synaptic input. The results of such experiments, together with the knowledge of the course and location of endings of the tympanal receptor fibers in the ventral cord, have been used to construct sufficient connectivity diagrams for the ascending ventral-cord neurons (Figs. 8, 12, 17, 22, 26). The following general properties have been revealed: 1. Between receptor cells and ascending auditory ventral-cord neurons, as a rule, interneurons are interposed; this accounts for the observation that the postsynaptic structures of the ventral-cord neurons are for the most part situated outside the auditory neuropile. 2. Six basic connectivity patterns underlie the responses of the 14 ventral-cord neurons described; that is, the same response patterns are produced in sets of two or three ventral-cord neurons, owing to synaptic connectivities involving the same interneurons. The distinctions between the response patterns of the 14 neurons are interpretable as stemming from the addition of another interneuron to, or by elimination of a facilitatory or inhibitory connection from, the basic connectivity patterns.