1. An interneurone specifically sensitive to substratum vibration was identified in the crayfish circumoesophageal connective. The interneurone, called B1 in this paper, received excitatory input from the statocysts on both sides. Electrical stimulation of the statocyst nerve elicited several spikes in the interneurone with latencies that depended on which side was stimulated.

  2. B1 responded phasically to artificial bending of the statocyst sensory hairs. The response was similar to that of the phasic-type receptor in the statocyst.

  3. The morphology of B1 was studied by an intracellular staining technique using nickel chloride and subsequent silver intensification. The interneurone projects its neurite arborization to the dorsal part of the deutocerebrum and parolfactory lobe on both sides, where the statocyst primary afferents also project. The overlapping of central projections, together with the properties of the response of B1; suggests that the interneurone receives excitatory input from the phasic-type receptors and transmits information about phasic body movement, but not static positional information, to the posterior ganglia.

  4. Branches of B1 also project to the antennal and tegumentary lobes ipsilateral to the axon. B1 may receive additional mechanosensory information from the cuticular sensory hairs on the antennae and the cephalic body surface.

The statocyst sensory nerve of crustaceans includes various functional types of sensory afferent (Cohen, 1955, 1960; Ozeki et al. 1978; Takahata and Hisada, 1979). In the crayfish Procambarus clarkii Girard, the afferents may be classified as being of either the tonic type or the phasic type according to the time course of their excitatory responses. Although most of them respond only to statocyst hair deflection towards the centre of the crescent, some of the phasic-type receptors respond transiently regardless of the direction of hair deflection (Takahata and Hisada, 1979).

In a preceding paper (Nakagawa and Hisada, 1989), we identified seven statocyst interneurones which responded tonically to hair deflection towards the centre of the crescent. We suggested that three of them would correspond to the statocyst interneurones identified physiologically by tilting experiments (Takahata and Hisada, 1982b). The tonic-type receptors are most likely to be functionally connected to these statocyst interneurones and convey positional information about body tilting. However, it is still unclear how the information from the phasic-type receptors is processed and represented in the central nervous system. In the crab Scylla serrata a semicircular canal interneurone, fibre 5, has been shown to carry not only directional low-frequency information but also high-frequency vibrational information (Fraser, 1975).

In this study, we report a descending interneurone which responds transiently to statocyst hair deflection regardless of the direction of deflection.

Animals

Adult crayfish, Procambarus clarkii Girard, of either sex, measuring 8–11 cm in length, were obtained commercially and kept in laboratory tanks before use.

Preparations

Both chelipeds and all walking legs were cut away and stomach contents were removed by suction with a pipette. For experiments involving mechanical stimulation, the animal was placed dorsal-side-up in a chamber filled with crayfish saline (Van Harreveld, 1936). A small portion of the dorsal carapace was removed to expose the circumoesophageal connectives. To stimulate the statocyst sensory hairs mechanically, the rostrum and the eyestalk on the side of the body to be stimulated were removed and the hairs covering the opening of the statocyst were then cut away. The statolith was washed out with a fine saline jet directed into the lumen. Finally, the chitinous dorsal covering over the statocyst was carefully removed to expose the statocyst floor and sensory hairs. The antennules were restrained by a rubber band to prevent them from moving during the experiment.

For experiments using electrical stimulation, the animals were placed ventral-side-up in a chamber filled with crayfish saline. The mouthparts and epistome were removed to expose the brain and circumoesophageal connectives. Then the antenna and the proximal joint of the antennule were cut away to expose the statocyst nerve.

Stimulation

Mechanical vibration of the substratum was achieved by tapping the side of the chamber using an acrylic rod attached to a solenoid driven by a waveform generator through a current driver. Statocyst hairs were stimulated using a loudspeaker, as previously described (Nakagawa and Hisada, 1989). Both types of mechanical stimulation were monitored by recording the current passing through the solenoid or the loudspeaker. For stimulation of the hairs, the monitor showed the time course and polarity of the stimulation, as in the previous paper (Nakagawa and Hisada, 1989). For tapping of the chamber, however, the monitor only showed that the acrylic rod had struck the side of chamber around the top of the ‘hill’ appearing on the monitoring trace. In later experiments, a stimulus intensity that was critical to elicit B1 spikes only was adopted, unless otherwise noted (Figs 3A,B, 4D, 5D). To determine the effective stimulus at this intensity, the substratum vibration was calibrated using a piezoelectric transducer (Fig. 1). The stimulus waveform was found to be a quickly damped oscillation of several successive wave peaks at about 125 Hz. The maximal amplitude of the stimulus vibration was found to be about 1.4 μm by extrapolation. Thus, except when specifically noted, the preparation was stimulated by this vibration regardless of the shape of the hill.

Fig. 1.

Actual vibratory movement of the substratum caused by the current method of stimulation. The upper trace is the recording from the medio-ventral small bundle of the circumoesophageal connective (CC). Only B1 spikes (dots) are elicited by tapping at 10 Hz. The middle trace shows the waveform of the substratum vibration recorded by a piezoelectric transducer (SV). The lower trace is the current applied to the solenoid. Throughout the following figures, this current profile was used as the stimulus monitor (SM).

Fig. 1.

Actual vibratory movement of the substratum caused by the current method of stimulation. The upper trace is the recording from the medio-ventral small bundle of the circumoesophageal connective (CC). Only B1 spikes (dots) are elicited by tapping at 10 Hz. The middle trace shows the waveform of the substratum vibration recorded by a piezoelectric transducer (SV). The lower trace is the current applied to the solenoid. Throughout the following figures, this current profile was used as the stimulus monitor (SM).

To stimulate the statocyst nerve electrically, the nerve was placed on a silver hook electrode enclosed in an oil-filled capillary. A square pulse of 0.1 ms duration was used for stimulation (Fig. 2A).

Fig. 2.

(A) Experimental arrangement for extracellular and intracellular recording of B[ spikes. The electrode for stimulation was placed on the statocyst nerve. Intracellular recordings were made from the circumoesophageal connective anterior to the extracellular recording site. (B) The response of interneurone B1 to electrical stimulation of the statocyst nerve on the side contralateral to the axon. In this preparation, Bj generated two spikes (arrows) in response to the stimulation. Note the 1:1 correspondence of the intracellular spikes with the extracellular ones.

Fig. 2.

(A) Experimental arrangement for extracellular and intracellular recording of B[ spikes. The electrode for stimulation was placed on the statocyst nerve. Intracellular recordings were made from the circumoesophageal connective anterior to the extracellular recording site. (B) The response of interneurone B1 to electrical stimulation of the statocyst nerve on the side contralateral to the axon. In this preparation, Bj generated two spikes (arrows) in response to the stimulation. Note the 1:1 correspondence of the intracellular spikes with the extracellular ones.

Fig. 3.

The responses of several descending interneurones to a soft tapping of the chamber at 10 Hz with decreasing amplitude (A–C). The upper record in each panel is the extracellular recording from the medio-ventral small bundle of the circumoesophageal connective (CC). The lower trace monitors the applied current (SM). In A and B, several descending interneurones respond to the stimulation, whereas in C only one descending interneurone, identified as interneurone B1; generates five spikes in response to the stimulation. The dots mark the B1 spikes.

Fig. 3.

The responses of several descending interneurones to a soft tapping of the chamber at 10 Hz with decreasing amplitude (A–C). The upper record in each panel is the extracellular recording from the medio-ventral small bundle of the circumoesophageal connective (CC). The lower trace monitors the applied current (SM). In A and B, several descending interneurones respond to the stimulation, whereas in C only one descending interneurone, identified as interneurone B1; generates five spikes in response to the stimulation. The dots mark the B1 spikes.

Recording and staining

The activity of the circumoesophageal connective was recorded extracellularly as previously described (Nakagawa and Hisada, 1989). To record the spike discharge from the interneurone B1, the circumoesophageal connective was split into a smaller bundle containing the axon so that spikes of this neurone could be clearly distinguishable from others.

In the ventral preparation, intracellular recordings were made from the small bundle of the circumoesophageal connective, using a glass microelectrode placed anterior to the extracellular recording site. The connective was stabilized on a platform. The microelectrode was filled with 0.25 mol l−1 NiCl2 (d.c. resistance, 30-80 MQ in saline) (Fig. 2A). The best recordings were obtained using electrodes having a d.c. resistance of 50–55 MΩ. Confirmation that the impaled neurone was the descending interneurone whose response was recorded extracel-lularly was given by the 1:1 correspondence between the extracellular and intracellular spikes (Fig. 2B).

After the response of the interneurone had been examined, nickel ions were injected into the interneurone by applying 10 nA depolarizing current pulses of 0.5 s in duration at 1 Hz for 1 h. After incubation at room temperature, the nickel was precipitated by rubeanic acid. Silver intensification was performed routinely on whole-mount specimens (Bacon and Altman, 1977). The intensified brain was dehydrated in alcohol and cleared in methyl salicylate. Drawings of the stained interneurones were made with a camera lucida. The description of the morphology was based on four successful stains of the interneurone.

Transverse paraffin sections, 10 μm thick, were made of the brain and the connective containing the stained neurone. Drawings were made with a camera lucida to reconstruct the entire structure of the interneurone and to examine its central projection. The nomenclature of the cell body clusters within the brain was taken from the scheme devised by Tautz and Tautz (1983).

A vibration-sensitive descending statocyst interneurone

Descending interneurones which responded to tapping of the experimental chamber were identified in the circumoesophageal connectives. An interneurone which was specifically sensitive to the mode of vibratory stimulation was found to receive excitatory input from statocysts on both sides (see below). We designated it as Bi, based on its bilateral input.

Several units including Bj responded to hard tapping. As the intensity of the tapping was decreased gradually, the number of responding units decreased until only interneurone Bi responded to the stimulus (Fig. 3A–C). We adopted this stimulus intensity for the following experiments. Removal of the statolith on the side ipsilateral to the interneurone axon had little effect on its response to tapping. However, when the contralateral one was subsequently washed away, the interneurone response disappeared completely (Fig. 4A–C). No response of the interneurone was observed after this bilateral statolith removal, even with a drastic increase in stimulus intensity (Fig. 4D). Similar results were obtained when statolith removal was performed in the reverse order, i.e. removal of the first (contralateral) statolith had little effect, whereas removal of the second (ipsilateral) statolith abolished the response (Fig. 5A–D). This observation indicated that the interneurone received vibratory input exclusively, but bilaterally, from the statocysts.

Fig. 4.

The effect of statolith removal on the response of interneurone to a soft tapping at 10 Hz. In each panel the upper four traces are extracellular recordings from the medio-ventral small bundle of the circumoesophageal connective (CC). The responses to four successive cycles of stimulation are shown. The lowest trace monitors the applied current (SM). In the top trace, Bi spikes are indicated by dots. (A) Both statocysts intact. (B) The statolith on the side ipsilateral to the axon of the interneurone is removed. Note the slightly reduced response of interneurone B1 (C) The contralateral statolith is also removed subsequently. The B1 response disappears completely. (D) The amplitude of stimulation is increased after removal of both statoliths. B1 never responds to the stimulation. All recordings are from the same animal.

Fig. 4.

The effect of statolith removal on the response of interneurone to a soft tapping at 10 Hz. In each panel the upper four traces are extracellular recordings from the medio-ventral small bundle of the circumoesophageal connective (CC). The responses to four successive cycles of stimulation are shown. The lowest trace monitors the applied current (SM). In the top trace, Bi spikes are indicated by dots. (A) Both statocysts intact. (B) The statolith on the side ipsilateral to the axon of the interneurone is removed. Note the slightly reduced response of interneurone B1 (C) The contralateral statolith is also removed subsequently. The B1 response disappears completely. (D) The amplitude of stimulation is increased after removal of both statoliths. B1 never responds to the stimulation. All recordings are from the same animal.

Fig. 5.

The effect of statolith removal in the reverse order to that in Fig. 4 on the response of interneurone BP The experimental conditions and data presentation are the same as those in Fig. 4. (A) Both statocysts intact. (B) The contralateral statolith is removed. Note the slight reduction in the Bi response. (C) The ipsilateral statolith is also removed subsequently. The Bj response almost disappears. (D) The amplitude of stimulation is increased after removal of both statoliths. B1 hardly responds to the stimulation.

Fig. 5.

The effect of statolith removal in the reverse order to that in Fig. 4 on the response of interneurone BP The experimental conditions and data presentation are the same as those in Fig. 4. (A) Both statocysts intact. (B) The contralateral statolith is removed. Note the slight reduction in the Bi response. (C) The ipsilateral statolith is also removed subsequently. The Bj response almost disappears. (D) The amplitude of stimulation is increased after removal of both statoliths. B1 hardly responds to the stimulation.

Difference in response between interneurones C1 and B1

Of the crayfish statocyst interneurones reported previously, interneurone C1 has been examined most thoroughly because of its large axon diameter (Takahata and Hisada, 1982a; Nakagawa and Hisada, 1989), and it can be considered as representative of the position-sensitive interneurones. The responses of interneurone Bi were compared with those of Q. Their responses to artificial bending of the statocyst hairs at various frequencies were examined first. For interneurone C1, the stimulation was applied only to the contralateral statocyst, since C1 is known to receive dominant input from the contralateral statocyst (Takahata and Hisada, 1982a). For interneurone Bh which receives bilateral input from the statocysts, stimulation was applied to each statocyst alternately.

Interneurone Ci showed a directional tonic response to stimulation at 0.1 Hz. However, this directionality became ambiguous at 1Hz, and there was no response to stimulation at 10 Hz (Fig. 6A). Further experiments showed that directional sensitivity was preserved well at 0.5 Hz, but disappeared at around 2Hz (Fig. 6B,C).

Fig. 6.

Responses of interneurone C1 to sensory hair deflection towards and away from the centre of the sensory crescent of the statocyst on the side contralateral to the axon. One cycle of the stimulus was applied at various frequencies. (A) The upper three traces are extracellular recordings from the whole circumoesophageal connective (CC). The first trace shows the response to the stimulation at 10Hz. The C1 spike is absent. The two large spikes are from interneurone C4. The second and third traces show the responses to the stimulation at 1 and 0.1 Hz, respectively. At 0.1 Hz, C1 shows a directional response. The lowest trace monitors current applied to the stimulation apparatus (SM). Upward excursion indicates inward movement of the statocyst hairs. In B and C, the upper traces are extracellular recordings from the whole circumoesophageal connective (CC). The lower traces monitor stimulus current (SM). The frequencies of stimulation are 0.5 and 2 Hz in B and C, respectively. Note that C1 preserves directional sensitivity at 0.5 Hz and that it disappears at 2 Hz. Calibration: top trace, 20 ms; second trace, 0.2 s; third trace, 2 s in A, 0.5 s in B, 0.1 s in C. All recordings were obtained from the same preparation.

Fig. 6.

Responses of interneurone C1 to sensory hair deflection towards and away from the centre of the sensory crescent of the statocyst on the side contralateral to the axon. One cycle of the stimulus was applied at various frequencies. (A) The upper three traces are extracellular recordings from the whole circumoesophageal connective (CC). The first trace shows the response to the stimulation at 10Hz. The C1 spike is absent. The two large spikes are from interneurone C4. The second and third traces show the responses to the stimulation at 1 and 0.1 Hz, respectively. At 0.1 Hz, C1 shows a directional response. The lowest trace monitors current applied to the stimulation apparatus (SM). Upward excursion indicates inward movement of the statocyst hairs. In B and C, the upper traces are extracellular recordings from the whole circumoesophageal connective (CC). The lower traces monitor stimulus current (SM). The frequencies of stimulation are 0.5 and 2 Hz in B and C, respectively. Note that C1 preserves directional sensitivity at 0.5 Hz and that it disappears at 2 Hz. Calibration: top trace, 20 ms; second trace, 0.2 s; third trace, 2 s in A, 0.5 s in B, 0.1 s in C. All recordings were obtained from the same preparation.

Interneurone B1 did not show any directional tonic response to stimulation at either 0.1 or 1 Hz. When a response was observed it was transient and occurred at the onset or the end of mechanical stimulation. Furthermore, unlike interneurone C15 interneurone Bt responded reliably to stimulation at 10Hz (Fig. 7). It showed a reliable response with a fixed latency to stimulation at 5 Hz, and a labile response to stimulation at 2 Hz (Fig. 8). Interneurone Bi showed similar responses to stimulation of the statocyst sensory hairs on either side.

Fig. 7.

Responses of interneurone Bi to statocyst sensory hair deflection towards and away from the centre of the sensory crescent at various frequencies. In A and B, the upper three traces are extracellular recordings from the medio-ventral small bundle of the circumoesophageal connective (CC). The uppermost trace shows the response to stimulation at 10 Hz. The second and third traces show the responses to stimulation at 1 and 0.1 Hz, respectively. The lowest trace monitors the stimulus current (SM). Upward excursion indicates inward movement of the sensory hairs. A and B show the response of Bi to contralateral and ipsilateral statocyst stimulation, respectively. Interneurone Bi shows similar responses to statocyst sensory hair stimulation on either side. Note that Bi reliably responds to stimulation at 10 Hz but not to stimulation at 1 and 0.1 Hz. In the first trace, all Bi spikes are indicated by dots, whereas in the second and third traces, only the first Bi spikes are indicated by dots. Calibration: first trace, 20ms; second trace, 0.2 s; third trace, 2 s. All recordings were obtained from the same preparation as those in Fig. 6.

Fig. 7.

Responses of interneurone Bi to statocyst sensory hair deflection towards and away from the centre of the sensory crescent at various frequencies. In A and B, the upper three traces are extracellular recordings from the medio-ventral small bundle of the circumoesophageal connective (CC). The uppermost trace shows the response to stimulation at 10 Hz. The second and third traces show the responses to stimulation at 1 and 0.1 Hz, respectively. The lowest trace monitors the stimulus current (SM). Upward excursion indicates inward movement of the sensory hairs. A and B show the response of Bi to contralateral and ipsilateral statocyst stimulation, respectively. Interneurone Bi shows similar responses to statocyst sensory hair stimulation on either side. Note that Bi reliably responds to stimulation at 10 Hz but not to stimulation at 1 and 0.1 Hz. In the first trace, all Bi spikes are indicated by dots, whereas in the second and third traces, only the first Bi spikes are indicated by dots. Calibration: first trace, 20ms; second trace, 0.2 s; third trace, 2 s. All recordings were obtained from the same preparation as those in Fig. 6.

Fig. 8.

(A–D) In each panel, the upper three traces are extracellular recordings from the medio-ventral small bundle of the circumoesophageal connective (CC). The responses to three successive stimuli (hair deflection) are shown. The lowest trace monitors the stimulus current (SM). The responses of interneurone Bi to contralateral statocyst stimulation at 5 and 2 Hz are shown in A and C, respectively. The responses to ipsilateral statocyst stimulation at 5 and 2 Hz are shown in B and D, respectively. Note that the interneurone shows reliable responses to statocyst stimulation on either side at 5Hz but not at 2Hz. In the uppermost trace, Bi spikes are indicated by dots. Calibration: 50 ms in A and B; 0.1 s in C and D. All recordings were obtained from the same preparation as those in Fig. 6.

Fig. 8.

(A–D) In each panel, the upper three traces are extracellular recordings from the medio-ventral small bundle of the circumoesophageal connective (CC). The responses to three successive stimuli (hair deflection) are shown. The lowest trace monitors the stimulus current (SM). The responses of interneurone Bi to contralateral statocyst stimulation at 5 and 2 Hz are shown in A and C, respectively. The responses to ipsilateral statocyst stimulation at 5 and 2 Hz are shown in B and D, respectively. Note that the interneurone shows reliable responses to statocyst stimulation on either side at 5Hz but not at 2Hz. In the uppermost trace, Bi spikes are indicated by dots. Calibration: 50 ms in A and B; 0.1 s in C and D. All recordings were obtained from the same preparation as those in Fig. 6.

This obvious difference in response between C1 and B1 suggests that they have different functional roles in statocyst information processing (see Discussion).

Responses of interneurone B1 to electrical stimulation of the statocyst nerve

To investigate the connection between interneurone B1 and the statocyst afferents, the statocyst nerve was electrically stimulated and the effect on the interneurone activity was examined.

The interneurone responded with a single spike to a single electrical stimulation (0.1ms duration) of either statocyst nerve at lower voltages. Increasing the stimulating voltage increased the number of spikes in the response (Fig. 9). The interneurone generated at most three or four spikes in response to this single stimulation at the supramaximal intensity. There was a tendency for more spikes to be elicited by ipsilateral statocyst nerve stimulation than by contralateral stimulation (Table 1, third spike). The first spike in response to electrical stimulation of either nerve occurred with a short fixed latency (Table 1; Fig. 10). However, the second and third spikes occurred with more variable latencies, as indicated by the high S.E. values (Table 1). The variability of the second and third spike latencies was less for ipsilateral stimulation than for contralateral stimulation (Table 1; Fig. 10).

Table 1.

Latencies of B1 spikes elicited by electrical stimulation to the contralateral and ipsilateral statocyst nerve

Latencies of B1 spikes elicited by electrical stimulation to the contralateral and ipsilateral statocyst nerve
Latencies of B1 spikes elicited by electrical stimulation to the contralateral and ipsilateral statocyst nerve
Fig. 9.

Responses of interneurone B1 to electrical stimulation of the ipsilateral statocyst nerve. The stimulus intensity was increased gradually (A, 3V; B, 4V; C, 6 V). Each panel shows the responses to three or four successive trials. Extracellular spikes of interneurone Bi (dots) were identified by a soft tapping beforehand. Increasing stimulus intensity both caused the short-latency response and subsequently recruited the long-latency responses.

Fig. 9.

Responses of interneurone B1 to electrical stimulation of the ipsilateral statocyst nerve. The stimulus intensity was increased gradually (A, 3V; B, 4V; C, 6 V). Each panel shows the responses to three or four successive trials. Extracellular spikes of interneurone Bi (dots) were identified by a soft tapping beforehand. Increasing stimulus intensity both caused the short-latency response and subsequently recruited the long-latency responses.

Fig. 10.

Histogram showing the latencies of B1 spike discharge in response to a single electrical stimulus applied to the ipsilateral (A) or contralateral (B) statocyst nerve. Each histogram is based on 50 stimulation tests in seven preparations. Stimulus intensity was 7 V throughout the experiment.

Fig. 10.

Histogram showing the latencies of B1 spike discharge in response to a single electrical stimulus applied to the ipsilateral (A) or contralateral (B) statocyst nerve. Each histogram is based on 50 stimulation tests in seven preparations. Stimulus intensity was 7 V throughout the experiment.

The short and fixed latency of the first spike suggests that it might be elicited monosynaptically in response to both ipsilateral and contralateral statocyst nerve stimulation. The long and variable latency of the following spikes indicates that they are elicited polysynaptically. Thus, the connections of interneurone B1 with the statocyst sensory neurone may be organized in parallel pathways similar to those of interneurone C1 (Takahata and Hisada, 1982a; see Discussion).

Morphology of interneurone B1

After physiological identification of B1 by the response described above, we studied its morphology by intracellular staining. Penetration of the interneurone was recognized by 1:1 correspondence between intracellular and extracellular spikes (Fig. 2). Four interneurones that had physiological properties matching those of Bi and were filled successfully showed the same morphology (Fig. 11).

Fig. 11.

Morphology of interneurone B1. (A) Dorsal view of the NiClj-injected interneurone. Arrow 1 indicates the characteristic thick secondary neurite (see text). The dotted line shows the midline of the ganglion. (B) Lateral view of the same interneurone as in A. The arrowhead indicates a gentle loop, and arrow 2 indicates the longest tertiary neurite (see text), ad indicate the positions of the sections reconstructed in Fig. 12. (C) Dorsal view of the brain at a lower magnification showing the position of the neuropile in the ganglion and the projecting area of Bj (stippled area). Op, optic lobe; Pa, parolfactory lobe; Ol, olfactory lobe; Ac, accessory lobe; An, antennal lobe. (D) Cross-section of the right circumoesophageal connective. The darkened rounded profile is the axon of the stained interneurone B1 (arrow), mg, medial giant fibre; lg, lateral giant fibre. In A–C, anterior is at the top. In B, dorsal is to the left. Calibration; 100pm in A and B, 50pm in D. B and D were obtained from the same preparation.

Fig. 11.

Morphology of interneurone B1. (A) Dorsal view of the NiClj-injected interneurone. Arrow 1 indicates the characteristic thick secondary neurite (see text). The dotted line shows the midline of the ganglion. (B) Lateral view of the same interneurone as in A. The arrowhead indicates a gentle loop, and arrow 2 indicates the longest tertiary neurite (see text), ad indicate the positions of the sections reconstructed in Fig. 12. (C) Dorsal view of the brain at a lower magnification showing the position of the neuropile in the ganglion and the projecting area of Bj (stippled area). Op, optic lobe; Pa, parolfactory lobe; Ol, olfactory lobe; Ac, accessory lobe; An, antennal lobe. (D) Cross-section of the right circumoesophageal connective. The darkened rounded profile is the axon of the stained interneurone B1 (arrow), mg, medial giant fibre; lg, lateral giant fibre. In A–C, anterior is at the top. In B, dorsal is to the left. Calibration; 100pm in A and B, 50pm in D. B and D were obtained from the same preparation.

Interneurone Bi has its cell body (approximately 30 μm in diameter for the major axis) in the anterior cluster on the side ipsilateral to the axon. The primary neurite forms a gentle loop at the deutocerebral region (Fig. 11B, arrowhead). The characteristic secondary neurite with an initial diameter of about 10 μm (Figs 11 A, 12c, arrow 1) arises from the primary neurite just anterior to the loop extends medio-ventrally across the midline, and terminates to branch in the parolfactory lobe on the side contralateral to the axon (Fig. 12b,c).

Fig. 12.

Drawings of transverse sections through interneurone Bj. Each section consists of three superimposed serial sections at intervals of 10–20 μm. The sections are viewed from the anterior end of the brain. The labelling (ad) of sections corresponds to that indicated in Fig. 11B. In a the longest tertiary neurite (arrow 2) does not terminate in the parolfactory lobe. In b the interneurone projects into the parolfactory lobe on both sides. In c the characteristic secondary neurite (arrow 1) extends across the midline. In d the interneurone projects into the antennal lobe ipsilateral to the axon. VPMC, ventral paired medial cluster; VUMC, ventral unpaired medial cluster; VPLC, ventral paired lateral cluster; DAC, dorsal anterior cluster; DMC, dorsal medial cluster; VUPC, ventral unpaired posterior cluster; VPPC, ventral paired posterior cluster; DPC, dorsal posterior cluster. Abbreviations for sensory neuropiles are the same as in Fig. 11C. Ventral is at the top.

Fig. 12.

Drawings of transverse sections through interneurone Bj. Each section consists of three superimposed serial sections at intervals of 10–20 μm. The sections are viewed from the anterior end of the brain. The labelling (ad) of sections corresponds to that indicated in Fig. 11B. In a the longest tertiary neurite (arrow 2) does not terminate in the parolfactory lobe. In b the interneurone projects into the parolfactory lobe on both sides. In c the characteristic secondary neurite (arrow 1) extends across the midline. In d the interneurone projects into the antennal lobe ipsilateral to the axon. VPMC, ventral paired medial cluster; VUMC, ventral unpaired medial cluster; VPLC, ventral paired lateral cluster; DAC, dorsal anterior cluster; DMC, dorsal medial cluster; VUPC, ventral unpaired posterior cluster; VPPC, ventral paired posterior cluster; DPC, dorsal posterior cluster. Abbreviations for sensory neuropiles are the same as in Fig. 11C. Ventral is at the top.

The secondary neurite gives rise to 9–10 orthogonal tertiary neurites (approximately 3 fim in diameter) which occur at equal distances (50–80 μm) along the secondary neurite. These tertiary neurites, except the longest one (Figs 11B, 12a, arrow 2), project into the parolfactory lobe on both sides (Fig. 11A). The longest one proceeds further towards the ventral side. The branching pattern of Bi in the deutocerebrum thus shows an orthogonal arrangement of secondary and tertiary neurites. This branching pattern resembles those of interneurones in the antennal lobe (Tautz and Tautz, 1983). After tapering of the primary neurite, interneurone Bi produces several secondary neurites (2–3 μm in diameter) which project into the tegumentary and antennal lobes and the dorsal part of the tritocerebrum ipsilateral to the axon. In the antennal lobe, an orthogonal arrangement can be seen, as in the parolfactory lobe.

The primary neurite becomes the axon entering the circumoesophageal connective, where it occupies a position in the ventro-medial quadrant (Fig. 11D, Wiersma’s area 72; Wiersma, 1958).

Several phasic descending statocyst interneurones have so far been described in the circumoesophageal connective of the crayfish Procambarus clarkii (Wiersma, 1958; Wiersma and Mill, 1965). However, the stimulation technique employed was not sensitive enough to determine specifically which receptors in the antennule were involved. In this study, we could identify the descending interneurone which responded phasically to statocyst hair stimulation. In the following sections we will discuss the functional and morphological differences between the phasic-and tonic-type descending statocyst interneurones by comparing interneurone B1 with the previously identified statocyst interneurone C1 (Takahata and Hisada, 1982a,b; Nakagawa and Hisada, 1989).

Range fractionation in the central representation of statocyst information

Interneurones C1 and B1 both respond to mechanical stimulation of the statocyst sensory hairs. However, their response frequency and input statocyst are quite different, indicating that they perform different functions.

The statocyst sensory nerves of crayfish and lobster include various types of sensory afferents. The first electrophysiological analysis of a single unit of this type, in Homaros americanus, revealed a variety of unit responses (Cohen, 1955). The position receptor gives a nonadapting and tonic response with a frequency depending on the absolute position of the sensory hair to which the unit is attached. The vibration receptor gives responses only to large acceleration or when the substratum is tapped. Recently, in the crayfish Orconectes limosas, vibration sensitivity of the statocyst has been investigated more quantitatively (Breithaupt and Tautz, 1988).

In the crayfish Procambarus clarkii, detailed study of the statocyst sensory neurones has revealed that they can be classified into two types of tonic and phasic cells according to their discharge patterns in response to sensory hair deflection (Takahata and Hisada, 1979). The tonic-type receptor showed an initial transient response when a specific hair was deflected, and the response was followed by a sustained spike discharge while the hair remained deflected. The phasic-type receptor showed only a transient excitatory response when a specific hair was deflected. It was also shown that most statocyst receptors could follow stimulus frequencies up to a maximum of about 5 Hz, but some could follow frequencies as high as 10 Hz.

Thus, the response properties of tonic-and phasic-type receptors are similar to those of interneurones C1 and B1 respectively. This suggests that the tonic-type receptors connect with the tonic-type descending statocyst interneurones such as C1, while the phasic-type receptors connect with the phasic-type interneurones such as B1.

Fraser (1975) showed in the crab that a semicircular canal interneurone, fibre 5, exhibited two peaks of sensitivity to oscillations of the antennules. He concluded that fibre 5 carried two sorts of information from the statocysts; low-frequency directional information and high-frequency vibrational information. In Procambarus clarkii, by contrast, the positional information and vibrational information are carried via two different channels of tonic-type and phasic-type statocyst interneurones, respectively.

Central connection between statocyst receptors and descending statocyst interneurones

In a previous study, it was shown that electrical stimulation of the contralateral statocyst caused both short-and long-latency responses in interneurone C1. As a result, it has been suggested that connection of interneurone C1 with the input statocyst sensory neurone is organized in a parallel way, one connection being monosynaptic and the other polysynaptic (Takahata and Hisada, 1982a). In the present study we have shown that electrical stimulation of the statocyst nerve on either side evoked several spikes with different latencies in interneurone B1 (Table 1). This result suggests that the input statocyst sensory neurone and interneurone B1 are connected via parallel pathways.

Although stimulation of the statocyst sensory hairs on both sides could activate interneurone B1; more spikes were elicited in response to ipsilateral stimulation than to contralateral stimulation, and the variability of the second and third spike latencies was less for ipsilateral stimulation (Table 1). The functional significance of this differential contribution from the two sides to activation of the interneurone is still to be investigated.

Relationship of morphology and function in two types of descending statocyst interneurones

Interneurone Bi projects to the dorsal part of the deutocerebrum and the parolfactory lobe on both sides. The statocyst primary afferents also project to these neuropile regions (Yoshino et al. 1983). The overlapping is consistent with the current report that interneurone B1 receives short-latency, probably monosynaptic, input from bilateral statocysts (Figs 9, 10). Interneurone B1 also extends its major branches to the antennal and tegumentary lobes on the ipsilateral side to the axon. It is known that the primary afferents connected to the antennal sensory hairs terminate in the antennal lobe (Taylor, 1975; Tautz and Tautz, 1983). It is also known that the tegumentary lobe that is part of a continuum with the antennal lobe is composed of axon terminals of mechanosensory afferents and interneurones whose receptive fields cover most of the animal’s surface (Kinnamon, 1979). This suggests that interneurone B1 also receives mechanosensory information from sensory hairs other than those of the statocysts. The threshold of nonstatocyst receptors for activating B1, however, was much higher than that of statocyst receptors (Figs 4, 5).

The projection pattern of interneurone B1 differs from that of the tonic-type statocyst interneurone (Nakagawa and Hisada, 1989) in two ways.

First, all seven tonic-type interneurones extend their major branches to the optic lobe where the fibres from the eyestalk ganglia terminate (Bethe, 1897; Hanstrom, 1925, 1948), while B1 lacks any branches in this neuropile. This difference in their morphology is consistent with physiological observations. It is well known that the equilibrium responses are controlled by complex interactions between the statocyst and visual input (Alverdes, 1926; Hisada, 1975; Takahata and Hisada, 1982a; Schöne et al. 1983; Neil, 1985). Moreover, it has also been reported that interneurone C1 receives visual input (Takahata and Hisada, 1982a). Tonic-type interneurones responsible for controlling equilibrium responses might well project to the optic lobe. In contrast, the lack of B1 branches in the optic lobe is also consistent with our conclusion that the phasic-type interneurone B1 is responsible for vibration reception rather than equilibrium responses.

Second, all seven tonic-type interneurones extend their major branches to the deutocerebrum either ipsilateral or contralateral to their axons, while B1 branches extend extensively and uniformly to the bilateral deutocerebrum. This appears to be consistent with the physiological finding that tonic-type interneurones receive their dominant input from one of the statocysts, whereas B1 receives input from both statocysts. However, we should note here that (1) the statocyst sensory afferents project to the deutocerebrum on both sides (Yoshino et al. 1983) and (2) the descending statocyst interneurones receive long-latency polysynaptic input from the input statocyst (Takahata and Hisada, 1982a; Figs 9, 10). These findings imply that the dendritic projection of statocyst interneurones in the deutocerebrum does not necessarily conform to the laterality of the input statocyst. In a previous report, indeed, we showed that all branches of interneurone S3 were restricted to the side ipsilateral to the axon, although this neurone received input from the contralateral statocyst (Nakagawa and Hisada, 1989). Thus, it is impossible to predict the input statocyst only from the projection patterns of an interneurone in the deutocerebrum.

Function of interneurone B1

Interneurone C1 and other tonic-type descending statocyst interneurones which show tonic and directional responses appear to be the principal channel for controlling equilibrium responses (Takahata and Hisada, 1982a; Nakagawa and Hisada, 1989). Interneurone B1, however, does not seem to be responsible for directly controlling the equilibrium responses since the interneurone shows only phasic responses and has no directional sensitivity. We conclude that interneurone B1 transmits information about contact or boundary vibration (Markl, 1983) rather than positional information. This conclusion raises the question of the role of vibrational information in controlling crayfish behaviour. Vibrational stimuli generally evoke flight reactions or reflex-like jumps directed away from the source of vibration. However, it is unlikely that the descending signals in B1 elicit these escape responses because the threshold for firing of the interneurone is very low.

One possible role of interneurone B1 could be to provide a general excitatory bias to the motor control system for behaviour which is affected by vibrational information – such as the escape response or target orientation (Tautz et al. 1981; Masters et al. 1982) – so that the following tonic signals could be readily integrated into the motor control signals. Alternatively, it is well known that contact or boundary vibrations are an important means of communication of sexual or aggressive interactions throughout the animal kingdom (Markl, 1983). Therefore, interneurone B, might participate in vibrational communication, although we know of no behavioural evidence for it in Procambarus clarkii. The role of Bi should be subjected to further investigation, especially at the behavioural level.

We thank Dr M. Takahata for his helpful comments on the manuscript. We are also grateful to Dr R. M. Glantz for his invaluable advice of the ventral-side-up brain preparation techniques. This work was supported by Grants-in-aid (nos 61480316 and 61840023) from the Japanese Ministry of Education, Science and Culture to MH.

Alverdes
,
F.
(
1926
).
State-, Photo-und Tangoreactionen bei zwei Garneelenarten
.
Z. vergl. Physiol
.
4
,
699
765
.
Bacon
,
J. P.
and
Altman
,
J. S.
(
1977
).
A silver intensification method for cobalt-filled neurons in wholemount preparations
.
Brain Res
.
138
,
359
363
.
Bethe
,
A.
(
1897
).
Das Nervensystem von Carcinus maenas, ein anatomischphysiologischer Versuch. I. Theil. II. Mittheil
.
Arch, mikrosk. Anat
.
50
,
589
639
.
Breithaupt
,
TH.
and
Tautz
,
J.
(
1988
).
Vibration sensitivity of the crayfish statocyst
.
Naturwissenschaften
75
,
310
312
.
Cohen
,
M. J.
(
1955
).
The function of receptors in the statocyst of the lobster Homarus americanus
.
J. Physiol., Lond
.
130
,
9
34
.
Cohen
,
M. J.
(
1960
).
The response patterns of single receptors in the crustacean statocyst
.
Proc. R. Soc. B
152
,
30
48
.
Fraser
,
P. J.
(
1975
).
Free hook hair and thread hair input to fiber 5 in the mud crab, Scylla serrata, during antennule rotation
.
J. comp. Physiol
.
103
,
291
313
.
Hanström
,
B.
(
1925
).
The olfactory centers in crustaceans
.
J. comp. Neurol
.
38
,
221
250
.
Hanström
,
B.
(
1948
).
The brain, the sense organs, and the incretory organs of the head in the Crustacea Malacostraca
.
Bull. biol. France et Belg. Suppl
.
33
,
98
126
.
Hjsada
,
M.
(
1975
).
Gravitational and visual control of eye movement in crayfish
.
Fortschr. Zool
.
23
,
162
173
.
Kinnamon
,
J. C.
(
1979
).
Tactile input to the crayfish tegumentary neuropile
.
Comp. Biochem. Physiol
.
63A
,
41
50
.
Markl
,
H.
(
1983
).
Vibration communication
.
In Neuroethology and Behavioral Physiology
(ed.
F.
Huber
and
H.
Markl
), pp.
332
353
.
Berlin, Heidelberg
:
Springer-Verlag
.
Masters
,
W. M.
,
Aicher
,
B.
,
Tautz
,
J.
and
Markl
,
H.
(
1982
).
A new type of water vibration receptor on the crayfish antenna. II. Model of receptor function
.
J. comp. Physiol
.
149
,
409
422
.
Nakagawa
,
H.
and
Hisada
,
M.
(
1989
).
Morphology of descending statocyst interneurons in the crayfish Procambarus clarkii Girard
.
Cell Tissue Res
.
255
,
539
551
.
Neil
,
D. M.
(
1985
).
Multisensory interactions in the crustacean’s equilibrium system
.
In Feedback and Motor Control in Invertebrates and Vertebrates
(ed.
W. J. P.
Barnes
and
M.
Gladden
), pp.
277
298
.
London
:
Croom Helm
.
Ozeki
,
M.
,
Takahata
,
M.
and
Hisada
,
M.
(
1978
).
Afferent response patterns of the crayfish statocyst with ferrite grain statolith to magnetic field stimulation
.
J. comp. Physiol
.
123
,
1
10
.
Schone
,
H.
,
Neil
,
D. M.
,
Scapini
,
F.
and
Dreissmann
,
G.
(
1983
).
Interaction of substrate, gravity and visual cues in the control of compensatory eye responses in the spiny lobster, Palinurus vulgaris
.
J. comp. Physiol
.
150
,
23
30
.
Takahata
,
M.
and
Hisada
,
M.
(
1979
).
Functional polarization of statocyst receptors in the crayfish Procambarus clarkii Girard
.
J. comp. Physiol
.
130
,
201
207
.
Takahata
,
M.
and
Hisada
,
M.
(
1982a
).
Statocyst interneurons in the crayfish Procambarus clarkii Girard. I. Identification and response characteristics
.
J. comp. Physiol
.
149
,
287
300
.
Takahata
,
M.
and
Hisada
,
M.
(
1982b
).
Statocyst interneurons in the crayfish Procambarus clarkii Girard. II. Directional sensitivity and its mechanism
.
J. comp. Physiol
.
149
,
301
306
.
Tautz
,
J.
,
Masters
,
W. M.
,
Aicher
,
B.
and
Markl
,
H.
(
1981
).
A new type of water vibration receptor on the crayfish antenna. I. Sensory physiology
.
J. comp. Physiol
.
144
,
533
541
.
Tautz
,
J.
and
Tautz
,
R. M.
(
1983
).
Antennal neuropile in the brain of the crayfish: Morphology of neurons
.
J. comp. Neurol
.
218
,
415
425
.
Taylor
,
R. C.
(
1975
).
Integration in the crayfish antennal neuropile: Topographic representation and multiple channel coding of mechanoreceptive submodalities
.
J. Neurobiol
.
6
,
475
499
.
Van Harreveld
,
A.
(
1936
).
A physiological solution for freshwater crustaceans
.
Proc. Soc. exp. Med
.
34
,
428
432
.
Wiersma
,
C. A. G.
(
1958
).
On the functional connections of single units in the central nervous system of the crayfish, Procambarus clarkii (Girard)
.
J. comp. Neurol
.
110
,
421
471
.
Wiersma
,
C. A. G.
and
Mill
,
P. J.
(
1965
).
“Descending” neuronal units in the commissure of the crayfish central nervous system; and their integration of visual, tactile and proprioceptive stimuli
.
I. comp. Neurol
.
125
,
67
94
.
Yoshino
,
M.
,
Kondoh
,
Y.
and
Hisada
,
M.
(
1983
).
Projection of statocyst sensory neurons associated with crescent hairs in the crayfish Procambarus clarkii Girard
.
Cell Tissue Res
.
230
,
37
48
.