Regional specialization in synaptic input and output in an identified local nonspiking interneuron of the crayfish revealed by light and electron microscopy

Neurons associated with the flip-flop activity in the lateral accessory lobe and ventral protocerebrum of the silkworm moth brain

The lateral accessory lobe (LAL) and the ventral protocerebrum (VPC) are a pair of symmetrical neural structures in the insect brain. The LAL-VPC is regarded as the major target of olfactory responding neurons as well as the control center for olfactory-evoked sequential zigzag turns. Previous studies of the silkworm moth Bombyx mori showed that these turns are controlled by long-lasting anti-phasic activities of the flip-flopping descending neurons with dendrites in the LAL-VPC. To elucidate the neural mechanisms underlying the generation of this alternating activity between the LAL-VPC units of both hemispheres, we first analyzed the detailed neural architecture of the LAL-VPC and identified five subregions. We then investigated the morphology and physiological responses of the LAL-VPC neurons by intracellular recording and staining and morphologically identified three types of bilateral neurons and three types of unilateral neurons. Bilateral neurons showed either brief or cyclic long-lasting responses. At least some neurons of the latter type produced gamma-aminobutyric acid (GABA). Unilateral neurons linking the LAL and VPC, in contrast, showed long-lasting or quick alternating activity. Timing analysis of the activity onset of each neural type suggests that quick reciprocal neural transmission between unilateral neurons would be responsible for the generation of long-lasting activity in one LAL-VPC unit, which lasts for up to a few seconds. Reciprocal inhibition and excitation by the bilateral neurons with long-lasting activities would mediate the alternating long-lasting activity between both LAL-VPC units, which might last for up to 20 seconds.

Effects of active conductance distribution over dendrites on the synaptic integration in an identified nonspiking interneuron

The synaptic integration in individual central neuron is critically affected by how active conductances are distributed over dendrites. It has been well known that the dendrites of central neurons are richly endowed with voltage- and ligand-regulated ion conductances. Nonspiking interneurons (NSIs), almost exclusively characteristic to arthropod central nervous systems, do not generate action potentials and hence lack voltage-regulated sodium channels, yet having a variety of voltage-regulated potassium conductances on their dendritic membrane including the one similar to the delayed-rectifier type potassium conductance. It remains unknown, however, how the active conductances are distributed over dendrites and how the synaptic integration is affected by those conductances in NSIs and other invertebrate neurons where the cell body is not included in the signal pathway from input synapses to output sites. In the present study, we quantitatively investigated the functional significance of active conductance distribution pattern in the spatio-temporal spread of synaptic potentials over dendrites of an identified NSI in the crayfish central nervous system by computer simulation. We systematically changed the distribution pattern of active conductances in the neuron's multicompartment model and examined how the synaptic potential waveform was affected by each distribution pattern. It was revealed that specific patterns of nonuniform distribution of potassium conductances were consistent, while other patterns were not, with the waveform of compound synaptic potentials recorded physiologically in the major input-output pathway of the cell, suggesting that the possibility of nonuniform distribution of potassium conductances over the dendrite cannot be excluded as well as the possibility of uniform distribution. Local synaptic circuits involving input and output synapses on the same branch or on the same side were found to be potentially affected under the condition of nonuniform distribution while operation of the major input-output pathway from the soma side to the one on the opposite side remained the same under both conditions of uniform and nonuniform distribution of potassium conductances over the NSI dendrite.

Functional significance of passive and active dendritic properties in the synaptic integration by an identified nonspiking interneuron of crayfish

Nonspiking interneurons control their synaptic output directly by membrane potential changes caused by synaptic activities. Although these interneurons do not generate spikes, their dendritic membrane is endowed with a variety of voltage-dependent conductances whose functional significance in synaptic integration remains unknown. We quantitatively investigated how the passive and active dendritic properties affect the synaptic integration in an identified nonspiking interneuron of crayfish by computer simulation using its multicompartment model based on electrophysiological measurements and three-dimensional morphometry. At the resting potential level, the attenuation factor (V(s)/V(t)) of a unitary synaptic potential in the course of its spread from a dendritic terminal (V(s)) to other terminals (V(t)) ranged from 4.42 to 6.30 with no substantial difference between hyperpolarizing and depolarizing potentials. The compound synaptic responses to strong mechanosensory stimulation could be reproduced in calculation only as the result of spatial summation of attenuated potentials, not as any single large potential. The characteristic response could be reproduced by assuming that the active conductances were distributed only in the dendritic region where the synaptic summation was carried out. The active conductances in other parts of the cell affected neither the shape of the compound synaptic response nor the dendritic spread of synaptic potentials. These findings suggest that the active membrane conductances do not affect the spatial distribution of synaptic potentials over dendrites but function in sculpting the summed synaptic potential to enhance temporal resolution in the synaptic output of the nonspiking interneuron.

Neural control mechanisms of the pheromone-triggered programmed behavior in male silkmoths revealed by double-labeling of descending interneurons and a motor neuron

Male silkmoths, Bombyx mori, exhibit a characteristic zigzagging behavior consisting of straight-line walking, zigzagging turns, and looping. The timing for shifting the turning direction is synchronized to the sideways head movements controlled by neck motor neurons (NMNs) including a cervical ventral NMN (cv1-NMN). It has been suggested that this programmed behavior is instructed by two types of activity patterns descending from the brain and the thoracic ganglion: one is a phasic excitation and the other is a state-dependent activity similar to the flipflop in electric memory circuits. These activities are shown by certain descending interneurons contained in two subsets of DNs, Group-I and -II DNs. However, it is not yet well understood which DNs are directly related to instructing this behavior. In order to understand neural control mechanisms of this programmed behavior, we investigated the morphological relationship between these DNs and the cv1-NMN, which is an index of this programmed behavior. We applied a double-labeling technique combining backfilling of the cv1-NMN and intracellular staining of single DNs. 3D confocal images revealed overlapping regions between the Group-I, -II DNs and the cv1-NMN. Group-IIA and -IID, which showed typical flipflop activities, Group-IIC DNs, which showed phasic excitation, and Group-IB DNs, which showed long-lasting inhibition had many overlapping regions on the cv1-NMNs. Our results indicate that the programmed behavior is instructed by these types of DNs.

Central inhibitory microcircuits controlling spike propagation into sensory terminals

The phenomenon of afferent presynaptic inhibition has been intensively studied in the sensory neurons of the chordotonal organ from the coxobasal joint (CBCO) of the crayfish leg. This has revealed that it has a number of discrete roles in these afferents, mediated by distinct populations of interneurons. Here we examine further the effect of presynaptic inhibition on action potentials in the CBCO afferents and investigate the nature of the synapses that mediate it. In the presence of picrotoxin, the action potential amplitude is increased and its half-width decreased, and a late depolarizing potential following the spike is increased in amplitude. Ultrastructural examination of the afferent terminals reveals that synaptic contacts on terminal branches are particularly abundant in the neuropil close to the main axon. Many of the presynaptic terminals contain small agranular vesicles, are of large diameter, and are immunoreactive for gamma-aminobutyric acid (GABA). These terminals are sometimes seen to make reciprocal connections with the afferents. Synaptic contacts from processes immunoreactive for glutamate are found on small-diameter afferent terminals. A few of the presynaptic processes contain numerous large granular vesicles and are immunoreactive for neither GABA nor glutamate. The effect that the observed reciprocal synapses might have was investigated by using a multicompartmental model of the afferent terminal.

Synaptic structure, distribution, and circuitry in the central nervous system of the locust and related insects

The Orthopteran central nervous system has proved a fertile substrate for combined morphological and physiological studies of identified neurons. Electron microscopy reveals two major types of synaptic contacts between nerve fibres: chemical synapses (which predominate) and electrotonic (gap) junctions. The chemical synapses are characterized by a structural asymmetry between the pre- and postsynaptic electron dense paramembranous structures. The postsynaptic paramembranous density defines the extent of a synaptic contact that varies according to synaptic type and location in single identified neurons. Synaptic bars are the most prominent presynaptic element at both monadic and dyadic (divergent) synapses. These are associated with small electron lucent synaptic vesicles in neurons that are cholinergic or glutamatergic (round vesicles) or GABAergic (pleomorphic vesicles). Dense core vesicles of different sizes are indicative of the presence of peptide or amine transmitters. Synapses are mostly found on small-diameter neuropilar branches and the number of synaptic contacts constituting a single physiological synapse ranges from a few tens to several thousand depending on the neurones involved. Some principles of synaptic circuitry can be deduced from the analysis of highly ordered brain neuropiles. With the light microscope, synaptic location can be inferred from the distribution of the presynaptic protein synapsin I. In the ventral nerve cord, identified neurons that are components of circuits subserving known behaviours, have been studied using electrophysiology in combination with light and electron microscopy and immunocytochemistry of neuroactive compounds. This has allowed the synaptic distribution of the major classes of neurone in the ventral nerve cord to be analysed within a functional context.

Quantitative analyses of anatomical and electrotonic structures of local spiking interneurons by three-dimensional morphometry in crayfish

We quantitatively investigated the three-dimensional structure of the dendrites of local spiking interneurons using a confocal laser scanning microscope in the terminal abdominal ganglion of crayfish. We also studied their passive membrane properties electrophysiologically using the single-electrode current clamp techniques to analyze their electrotonic structure. All of the local spiking interneurons examined in this study lacked distinctive axonal structure and had a monopolar cell body that was connected with a fine primary process to a thick main segment. Numerous fine secondary processes projected from the main segment in the ganglionic neuropile. The average anatomical length of a secondary process from the main segment to its terminal was 261.9 +/- 15.2 microm. The average input resistance and membrane time constant of local spiking interneurons, obtained from their voltage responses to intracellular injection of step current pulses in the main segment, were 15.2 +/- 1.6 MOmega and 13.9 +/- 1.9 msec, respectively. Calculation of the electrotonic length of dendritic processes based on morphological and physiological data obtained in this study revealed that the average electrotonic length of secondary processes in local spiking interneurons was significantly longer than in local nonspiking interneurons, although both types of local interneurons showed apparently similar anaxonic structure. The steady-state voltage attenuation factors for the secondary processes of local spiking interneurons were significantly greater than those of local nonspiking interneurons in both centrifugal and centripetal directions. The larger electrotonic structure of local spiking interneurons compared to that of nonspiking interneurons appears to be compensated for by their excitable dendritic membrane.

Electrophysiological and theoretical analysis of depolarization-dependent outward currents in the dendritic membrane of an identified nonspiking interneuron in crayfish

Depolarization-dependent outward currents were analyzed using the single-electrode voltage clamp technique in the dendritic membrane of an identified nonspiking interneuron (LDS interneuron) in situ in the terminal abdominal ganglion of crayfish. When the membrane was depolarized by more than 20 mV from the resting potential (65.0 +/- 5.7 mV), a transient outward current was observed to be followed by a sustained outward current. Pharmacological experiments revealed that these outward currents were composed of 3 distinct components. A sustained component (I(s)) was activated slowly (half rise time > 5 msec) and blocked by 20 mM TEA. A transient component (I(t1)) that was activated and inactivated very rapidly (peak time < 2.5 msec, half decay time < 1.2 msec) was also blocked by 20 mM TEA. Another transient component (I(t2)) was blocked by 100 microM 4AP, activated rapidly (peak time < 10.0 msec) and inactivated slowly (half decay time > 131.8 msec). Two-step pulse experiments have revealed that both sustained and transient components are not inactivated at the resting potential: the half-maximal inactivation was attained at -21.0 mV in I(t1), and -38.0 mV in I(t2). I(s) showed no noticeable inactivation. When the membrane was initially held at the resting potential level and clamped to varying potential levels, the half-maximal activation was attained at -36.0 mV in I(s), -31.0 mV in I(t1) and -40.0 mV in I(t2). The activation and inactivation time constants were both voltage dependent. A mathematical model of the LDS interneuron was constructed based on the present electrophysiological records to simulate the dynamic interaction of outward currents during membrane depolarization. The results suggest that those membrane conductances found in this study underlie the outward rectification of the interneuron membrane as well as depolarization-dependent shaping of the excitatory synaptic potential observed in current-clamp experiments.

Quantitative analyses of anatomical and electrotonic structures of crayfish nonspiking interneurons by three-dimensional morphometry

The three-dimensional structure of premotor nonspiking interneurons in the terminal abdominal ganglion of crayfish have been studied quantitatively by using a confocal laser-scanning microscope. Their passive membrane properties have also been studied electrophysiologically to analyze their electrotonic structure. In either one of the two major morphological types, anterolateral (AL) and posterolateral (PL), that are characterized by different locations of cell bodies in the ganglion, the monopolar cell body is connected with a fine primary process to a thick main segment projecting numerous fine secondary processes. These two types of cells share a common dendritic field in the neuropil, showing similar anatomical characteristics of dendrites. Electrotonic analyses based on the present anatomical and physiological measurements have revealed that the steady-state voltage-attenuation factors for the secondary processes were not statistically different between the AL- and PL-type cells. Comparison between the premotor nonspiking interneurons and an identified sensory nonspiking interneuron, which was studied previously, has revealed that voltage attenuation over secondary processes in both the centripetal and the centrifugal directions was significantly greater in the sensory than in the premotor interneurons, although the anatomical length of each secondary process from its terminal to the main segment was not different between them. Differences in the electrotonic structure between sensory and premotor nonspiking interneurons indicate their different modes of synaptic integration in the control of postsynaptic nerve cells.

Distribution of centrifugal neurons targeting the soma clusters of the olfactory midbrain among decapod crustaceans

To determine the distribution of two systems of centrifugal neurons innervating the soma clusters of the olfactory midbrain across decapod crustaceans, brains of the following nine species comprising most infraorders were immunostained with antibodies against dopamine and the neuropeptides substance P and FMRFamide: Macrobrachium rosenbergii, Homarus americanus, Cherax destructor, Orconectes limosus, Procambarus clarkii, Astacus leptodactylus, Carcinus maenas, Eriocheir sinensis and Pagurus bernhardus. One system consisting of several neurons with dopamine-like immunoreactivity that originate in the eyestalk ganglia was present in the four crayfish but not in any other species. These neurons project mainly into the lateral soma clusters (cluster 10) comprising the somata of ascending olfactory projection neurons and innervate very sparsely the medial soma clusters (clusters 9 and 11) containing the somata of local interneurons. In the innervation pattern of the lateral cluster, the dopamine-immunoreactive neurons showed large species-specific differences. The other system comprises a pair of giant neurons with substance P-like immunoreactivity. These neurons have somata in the median protocerebrum of the central brain and major projections into the lateral clusters and the core of the olfactory lobes, the neuropils that are the first synaptic relay in the central olfactory pathway of decapods; minor arborizations are present in the medial clusters. The system of substance P-immunoreactive giant neurons was present and of great morphological similarity in all studied species. Only in one species, the shrimp Macrobrachium rosenbergii, evidence for co-localization of FMRFamide-like with substance P-like immunoreactivity in these neurons was obtained. These and previously collected data indicate that the centrifugal neurons with dopamine-like immunoreactivity may be associated with the presence of an accessory lobe, a second-order neuropil that receives input from the olfactory lobe and only occurs in spiny lobsters, clawed lobsters and crayfish. The pair of centrifugal giant neurons with substance P-like immunoreactivity, on the other hand, appears to be a constitutive component of the decapod crustacean brain that most likely is functionally associated with the olfactory lobe. Both systems apparently exert modulatory functions on olfactory information processing by preferentially targeting the somata of the projection neurons. Thus, in the olfactory projection neurons, the somata seem to be more directly involved in information processing than in most other neurons of the arthropod CNS.

Combining laser scanning confocal microscopy and electron microscopy in studies of the insect nervous system

Experimentally determining the synaptic interconnections between neurons in the nervous system is laborious and difficult in any animal species, but especially so in many invertebrates, including insects, where neurons generally have large, finely branching neuritic trees that form both pre- and postsynaptic specializations in dense neuropils with other neuritic trees. Electron microscopy is needed to identify synapses, but correlation of synapse type and location with the overall branching patterns of neurons, which are visible readily only in the light microscope or through extensive reconstruction of serial electron-microscope sections, is very difficult. In this paper, we present a simple method that we have developed (Sun et al. (1995) J. Histochem. Cytochem., 43: 329-335) that combines laser scanning confocal microscopy and electron microscopy for the study of synaptic relationships of neurons in the antennal lobe, the first central neuropil in the olfactory pathway, of the moth Manduca sexta. Briefly, neurons are labeled by intracellular injection with neurobiotin or biocytin, and then processed with a gold-particle tag for electron microscopic study and a fluorescent tag for confocal microscopy, and embedded in plastic. The fluorescence of the labeled neuron in the plastic blocks is imaged in three dimensions with laser scanning confocal microscopy and then the neuron is thin-sectioned at precisely chosen depths for electron microscopic study. The fluorescence pattern can be monitored repeatedly between episodes of thin-sectioning, and subtraction of a fluorescence image from the previous fluorescence image reveals which fluorescent processes have been sectioned. In this way, electron microscopic detail can be mapped onto a three-dimensional light microscopic image of the neuron.

Ultrastructure of the circuit providing input to the crayfish lateral giant neurons

Labeled or otherwise identified neurons of the crayfish lateral giant escape reaction circuit were examined electron microscopically and the findings compared to expectations from physiology. Terminals of primary afferents contained clear, approximately 45 nm, irregularly round synaptic vesicles, while sensory interneuron terminals had slightly larger, 50 nm, more strictly round vesicles, permitting tentative classification based on anatomical criteria. Excitatory synapses on the lateral giants, believed from physiology to be electrical, generally had some gap junctions, but these were almost invariably paralleled by more prominent chemical junctional regions of unknown function. There may also be a class of interneurons making purely chemical synapses on the lateral giants. Synapses from primary afferents to sensory interneurons, believed from physiology to be cholinergic, had purely chemical morphology. Synapses with narrow elongated vesicles, similar to GABAergic vesicles seen in other neurons, frequently occurred on terminals of primary afferents. These synapses provide a basis for known presynaptic inhibition of afferent input. Consistent with physiology, such inhibitors sometimes also contacted the postsynaptic targets of the primary afferents and sometimes received input from other primary afferents. Afferent terminals also received some input from profiles rich in large dense cored vesicles. Presumptive inhibitory input found on proximal dendrites of lateral giants provides a basis for known recurrent inhibition. However, similar inhibitory synapses that sometimes received local input from excitors of the lateral giants were also found distally mixed with excitatory inputs. These provide a basis for recently discovered distal inhibitory input following excitation and for tonic inhibition.

Distribution and structure of identified tonic and phasic synapses between L-neurones in the locust ocellar tract

The ultrastructures and distributions of the discrete anatomical synapses which constitute two distinct types of output connections made by individual ocellar L-neurons, L1-3, are described. Outputs to neurones L4-5 are excitatory and transmit tonically, whereas reciprocal connections among the three L1-3 neurones are inhibitory and incapable of transmission for longer than a few milliseconds. The tonically transmitting synapses are located in the lateral ocellar tract and are made between the axons of L1-3, which do not receive inputs, and short branches of L4-5, which make no outputs. Each excitatory connection is composed of a few hundred discrete anatomical synapses, each characterised by a bar-shaped presynaptic density which is 0.15-1.5 microns in length and associated with a large number of round synaptic vesicles. Two postsynaptic profiles are apposed to each presynaptic density. Associated with tonic synapses are abundant invaginations of the presynaptic membrane. Synapses of the reciprocal, inhibitory, phasic connections occur in the protocerebral arbors of L1-3, among numerous output synapses of these neurones. Each phasic connection is composed of a few tens of discrete anatomical synapses. Each bar-shaped presynaptic density is associated with two postsynaptic profiles, and is 0.1-1.0 microns long. Compared with the tonic, excitatory connection, there are fewer vesicles and fewer invaginations of the presynaptic membrane associated with each synapse.

Long-lasting excitation of protocerebral bilateral neurons in the pheromone-processing pathways of the male moth Bombyx mori

Intracellular recording and staining with Lucifer yellow were used to characterize the responses and structure of pheromone-processing bilateral neurons in the protocerebrum of the brain of the male silkworm moth Bombyx mori. Numerous olfactory bilateral neurons innervated a particular neuropil region lateral to the central body, the lateral accessory lobe (LAL). The LALs are linked to each other by bilateral neurons with arborizations in each LAL. The LAL appears to be important for collecting the olfactory information from both sides of the brain. Many of the bilateral neurons showed a characteristic long-lasting excitation (LLE) that outlasted the olfactory stimuli (1.5 s). In some preparations, the LLE lasted more than 20 s and the firing gradually decreased to the background level.

Antennular projections to the midbrain of the spiny lobster. I. Sensory innervation of the lateral and medial antennular neuropils

The organization of sensory afferents in the antennular nerve (AN) of the spiny lobster and the central arborization of the afferents in the lateral and medial antennular neuropils (LAN, MAN) were analyzed by backfilling the AN with biocytin. The MAN receives primarily thick afferents (diameter greater than or equal to 10 microns) with a consistent pattern of arborization from the medial of the three major divisions of the AN. The LAN, in contrast, receives many thin to medium-sized afferents (diameter less than or equal to 0.3-5 microns), in addition some with diameters greater than or equal to 5 microns, from the lateral and dorsal divisions of the AN. In contrast to the consistent pattern of arborization in the MAN, afferents projecting to the LAN arborize in widely different patterns. Serially arranged, orthogonal side branches that are suggestive of topographical representation of the serially arranged sensilla on the antennule contribute to the stratification of the LAN. Together with existing electrophysiological data, these morphological findings are consistent with the idea that the MAN receives primarily mechanosensory (largely statocyst) input, as previously thought, but that the LAN receives chemosensory as well as mechanosensory input. The chemosensory input to the LAN would represent a novel pathway for processing chemosensory input from the antennule.

Antennular projections to the midbrain of the spiny lobster. II. Sensory innervation of the olfactory lobe

The projection pattern of antennular sensory afferents in the olfactory lobe (OL) of the spiny lobster, Panulirus argus, was examined by backfilling axons in the antennular nerve (AN) with biocytin. Thin, presumptive olfactory afferents from the lateral division of the AN form a tract in the brain that diverges into a dense plexus that completely envelops the glomerular cortex of the OL. Most of the thin (diameter less than or equal to 0.3-1 microns) afferents project to single glomeruli. About 10% of the thin afferents, however, branch in the plexus and project to multiple glomeruli. A smaller number of medium-sized to thick (diameter 2-10 microns), presumably mechanosensory, afferents also innervate the OL and co-project to multiple glomeruli with the thin afferents. Afferents arborize profusely within the columnar glomeruli into very fine processes that penetrate to the base of the columns, but selectively terminate in either the cap/subcap region or in the innermost part of the base of the columns, often with conspicuous terminal boutons, forming two distinct regions of presumptive synaptic output. These results suggest that 1) The majority of the OL innervation is provided by olfactory sensilla (aesthetascs), but that other types of sensilla provide additional, likely mechanosensory, input to the OL. 2) The projection of olfactory afferents is not strictly uniglomerular. 3) The columnar organization of crustacean olfactory glomeruli is functionally significant and may provide an evolutionary correlate of the recently proposed subdivision of the vertebrate olfactory bulb into "functional columns."

Physiology and morphology of protocerebral olfactory neurons in the male moth Manduca sexta

1. We have used intracellular recording and staining with Lucifer Yellow, followed by reconstruction from serial sections, to characterize the responses and structure of olfactory neurons in the protocerebrum (PC) of the brain of the male sphinx moth Manduca sexta. 2. Many olfactory protocerebral neurons (PCNs) innervate a particular neuropil region lateral to the central body, the lateral accessory lobe (LAL), which appears to be important for processing olfactory information. 3. Each LAL is linked by its constituent neurons to the ipsilateral lateral PC, where projection neurons from the antennal lobe terminate, as well as to other regions of the PC. The LALs are also linked to each other by bilateral neurons with arborizations in each LAL. 4. Some PC neurons showed long-lasting excitation (LLE) that outlasted the olfactory stimuli by greater than or equal to 1 s, and as long as 30 s in some preparations. LLE was more frequently elicited by the sex-pheromone blend than by individual pheromone components. All bilateral neurons that showed LLE had arborizations in the LALs. LLE responses were also recorded in a single local neuron innervating the mushroom body. 5. In some other PC neurons, pheromonal stimuli elicited brief excitations that recovered to background firing rates less than 1 s after stimulation.

Distribution and morphology of synapses on nonspiking local interneurones in the thoracic nervous system of the locust

The structure and distribution of synapses on nonspiking local interneurones in the metathoracic ganglion of the locust was revealed by electron microscopy following intracellular injection of horseradish peroxidase (HRP). Before staining, each interneurone was characterized physiologically as nonspiking and its output effects on motor neurones innervating muscles in a hindleg were investigated. Three nonspiking interneurones of different morphologies, each typical of a previously described population, were selected for detailed study. The first has a dorsal soma and ipsilateral neuropilar branches, the second a ventral soma and ipsilateral branches, and the third a ventral soma and contralateral branches. The somata have few trophospongial invaginations, and most of their volume is occupied by the nucleus. The initial parts of the primary neurites are either wrapped in glia or isolated in tracts from the neuropile and thus do not participate in synaptic interactions. Some of the larger secondary neurites are also wrapped in glia, but others both make and receive synaptic contacts. Output synapses have an array of some 500-1,600 round, agranular vesicles (diameter 47.0 +/- 5.7 nm; mean +/- S.D., n = 97) associated with a bar-shaped presynaptic density up to 0.3 micron long. Two postsynaptic processes, whose diameter can vary greatly, are usually associated with each presynaptic density. Processes making input synapses onto nonspiking local interneurones typically contain round, agranular vesicles and often make several contacts within a few microns. Serial reconstructions from one of the interneurones revealed input and output synapses intermingled on the larger processes with outputs dominating by a factor of 3:1, whereas on some of the thinner processes only input synapses are present. In the other two interneurones, however, both input and output synapses are present on the fine branches. No feature of the structure or distribution of synapses observed here on the nonspiking local interneurones distinguishes them from spiking neurones in the same ganglia.