Mechanosensory Input Research Articles

Mechanosensory afferent input and neuronal firing properties in rodent spinal laminae III–V: re-examination of relationships with analysis of responses to static and time-varying stimuli

Relationships between neuronal firing pattern and mechanosensory input in the deep dorsal horn were investigated using whole-cell recordings from isolated hamster spinal cord with innervation from an attached skin patch. Neurons that fired repetitively to depolarizing current (tonic cells) responded to both moving and static stimulation of their cutaneous receptive fields, and discharged continuously for the duration of stimulus application. Neurons responding to depolarizing current with transient, rapidly adapting firing (phasic cells) were significantly more responsive to stimulus movement than to static skin contact. Phasic cells typically issued a brief discharge at the onset or termination of a stimulus; their responses during static skin contact were weaker than tonic cells. Tonic cells were activated during both ramp and steady-state skin indentations, whereas phasic cells responded with their strongest excitation to displacement velocities exceeding 8 μm/ms. Mechanosensory input to phasic cells originated primarily from low threshold receptors, whereas tonic cells demonstrated a mixture of inputs from both low and high threshold sources. A third class of neurons responded to depolarizing current with a pronounced firing delay and displayed a sensitivity to cutaneous stimuli that was similar to tonic cells except they showed a modest decrease in firing as skin indentation velocity increased. The results suggest a correlation between functional properties of mechanoreceptive afferent fibers and intrinsic discharge properties of laminae III–V neurons that may significantly influence integration of cutaneous mechanosensory information at the first spinal relay.

Projection and local interneurons in the sixth abdominal ganglion of the sand crab Emerita analoga (Hippidae)

Hippid crabs are adapted for life in the surf zone of exposed sandy beaches, and their tailfan differs from the tailfans of other crustaceans with respect to morphology and motor control and in having nonspiking stretch receptors (NSR). To investigate how these crabs' mechanosensory systems are adapted to this turbulent environment, I used axonal back-filling and intracellular recording with dye-filled microelectrodes to describe afferent projections from the telson and morphologies and physiological responses of intersegmental and local neurons in abdominal ganglion 6 (A6) in Emerita analoga, then compared them with descriptions of corresponding neurons in A6 of crayfish. The density of afferent terminals and the proportion of projection neuron somata is lower in anterior A6 of E. analoga than in crayfish, perhaps correlated with a reduction in hydrodynamic setae. Many interneurons responded to afferent nerve stimulation and displayed activity correlated with levels of A6 motor output. NSR stretch depolarizes unilateral local neurons and terminals of axons entering A6 from the connective and hyperpolarizes bilateral local and projection neurons. The timing and duration of this inhibition would suppress mechanosensory input from the telson during uropod beating (homologue of crayfish's nongiant tail-flipping). Suppression of reafference during uropod beating may have been pivotal for evolution of hippids' ability to move rapidly across the water-sand interface in the slosh zone of sandy beaches. Homologies between A6 neurons in E. analoga and crayfish, suggested by morphological and physiological similarities, indicate that the NSRs connect to a neuronal network regulating exteroceptive input that was inherited from their tail-flipping ancestors.

Airway Stretch Receptor Behavior During Exposure to Prolonged Positive End-Expiratory Pressure

PURPOSE: Positive end-expiratory pressure (PEEP) is commonly used during mechanical ventilation in several clinical settings. Such a change in mechanical environment will affect the mechanosensory input from pulmonary slowly adapting receptors (SARs), altering cardiopulmonary function. However, we know little about how SARs behave during prolonged, graded PEEP application.

Modulation of mechanosensory responses by motoneurons that regulate skin surface topology in the leech.

Central regulation of somatosensory signals has been extensively studied, but little is known about their regulation in the periphery. Given the widespread exposure of the skin sensory terminals to the environment, it is of interest to explore how somatosensory sensitivity is affected by changes in properties of the skin. In the leech, the annuli that subdivide the skin can be erected under the control of the annulus erector (AE) motoneurons. To analyze whether this surface change influences mechanosensory sensitivity, we studied the responses of low threshold mechanosensory T cells to mechanical stimulation of the skin as AE motoneurons were activated. In segments of the body wall connected to the corresponding ganglion and submerged in an aqueous environment, T cells responded to localized bubbling on the skin and to water flow parallel to its surface. Excitation of AE motoneurons diminished these responses in a way that depended on the motoneuron firing frequency. Video recordings established that the range of AE firing frequencies that produced effective annulus erection coincided with that influencing T cell responses. In isolated ganglia, AE firing had no effect on T cell excitability, suggesting that annulus erection diminished T cell responsiveness to mechanical input. Counteracting this effect, mechanosensory inputs inhibited AE motoneurons. However, because depolarization of AE cells caused a decrease in their input resistance, the more active the motoneuron, the less sensitive it became to inhibitory signals. Thus when brought to fire, AE motoneurons would stay "committed" to a high activity level, and this would limit sensory responsiveness to incoming mechanical signals.

Enhanced Locomotion Caused by Loss of the Drosophila DEG/ENaC Protein Pickpocket1

Coordination of rhythmic locomotion depends upon a precisely balanced interplay between central and peripheral control mechanisms [1]. Although poorly understood, peripheral proprioceptive mechanosensory input is thought to provide information about body position for moment-to-moment modifications of central mechanisms mediating rhythmic motor output [2]. Pickpocket1 (PPK1) is a Drosophila subunit of the epithelial sodium channel (ENaC) family displaying limited expression in multiple dendritic (md) sensory neurons tiling the larval body wall and a small number of bipolar neurons in the upper brain [3]. ppk1 null mutant larvae had normal external touch sensation and md neuron morphology but displayed striking alterations in crawling behavior. Loss of PPK1 function caused an increase in crawling speed and an unusual straight path with decreased stops and turns relative to wild-type. This enhanced locomotion resulted from sustained peristaltic contraction wave cycling at higher frequency with a significant decrease in pause period between contraction cycles. The mutant phenotype was rescued by a wild-type PPK1 transgene and duplicated by expressing a ppk1RNAi transgene or a dominant-negative PPK1 isoform. These results demonstrate that the PPK1 channel plays an essential role in controlling rhythmic locomotion and provide a powerful genetic model system for further analysis of central and peripheral control mechanisms and their role in movement disorders.

A Lateral Excitatory Network in the Escape Circuit of Crayfish

A phasic stimulus directed to the rear of a crayfish (Procambarus clarkii) creates mechanosensory input to the lateral giant (LG) interneuron, a command neuron for escape. A single LG spike is necessary and sufficient to produce a highly stereotyped tail flip that thrusts the animal away from the source of stimulation. Here we describe a lateral excitatory network among primary afferent axons in the last abdominal ganglion of crayfish that produces nonlinear amplification of the sensory input to the command circuitry for escape. The lateral excitation is mediated by electrical synapses between central terminals of primary mechanosensory afferents. The network enables stimulated afferents to recruit unstimulated afferents that contribute additional input to LG and to mechanosensory interneurons that also converge on LG. When depolarized, the LG neuron increases its own inputs from primary afferents and primary interneurons by facilitating the recruitment of both. Conversely, hyperpolarization of LG reduces the excitability of primary afferents and primary interneurons. The crayfish's decision to escape, previously thought to lie exclusively in the synaptic integrative properties of LG, is now seen to depend on the interactions between LG dendritic postsynaptic potentials and the responses of primary afferent terminals in the lateral excitatory network.

Peripheral synapses and giant neurons in whip spiders

Among invertebrates the synapses between neurons are generally restricted to ganglia, i.e., to the central nervous system (CNS). As an exception, synapses occur in the sensory nerves of arachnid legs, indicating that some nervous integration is already taking place far out in the periphery. In the antenniform legs of whip spiders (Amblypygi), a very special synaptic circuit is present. These highly modified legs contain several large interneurons (giant neurons) that receive mechanosensory input from 700-1,500 tarsal bristles. Some of the sensory cell axons contact a giant neuron at its short, branched dendrite, a few at the soma, but most synapse onto the long giant axon. The fine structure of these synapses resembles that of typical chemical synapses in other arthropods. Although thousands of sensory fibers converge on a single giant neuron, there is no reduction in the actual number of sensory fibers, because these afferent fibers continue their course to the CNS after having made several en passant synapses onto the giant neuron. Touching a single tarsal bristle is sufficient to elicit action potentials in a giant neuron. Owing to the large diameter of the giant axon (10-20 microm), the action potentials reach the CNS within 55 ms, at conduction velocities of up to 7 m/s. However, mechanical stimulation of the tarsal bristles does not elicit a fast escape response, in contrast to giant fiber systems in earthworms, certain insects, and crayfishes. A quick escape is observed in whip spiders, but only after stimulation of the filiform hairs (trichobothria) on the regular walking legs. Although the giant fiber system in the antenniform legs undoubtedly provides a fast sensory pathway, its biological significance is not clearly understood at the moment.

Colonic mechanoreceptor inputs to rat lumbo-sacral dorsal horn neurones: distribution, thresholds and chemosensory modulation.

In order to better understand the central processing of visceral sensory information, we studied the responses of lumbo-sacral dorsal horn (L4-S1) neurones to colonic stimuli in anaesthetized rats. Twenty-four neurones responded to distal colonic distension with a 2.5-cm balloon; six of these were tested with proximal colonic distension, to which none responded. All neurones tested responded to somatic non-noxious inputs (tail movement). Responses to colonic distension were excitatory (n=22) or inhibitory (n=2). Sixteen neurones responded at a threshold of 20 mmHg or less, five at 20-40 mmHg, and three at 40-80 mmHg. Three of 10 neurones tested showed increased responses to colonic distension after intraluminal perfusion with bile. Bile itself did not evoke a response. We conclude that lumbo-sacral spinal neurones selectively receive mechanosensory inputs from the distal colon. Neurones respond at thresholds within and above the physiological range. Dorsal horn neurones receiving colonic mechanosensory inputs are not directly modulated by chemosensory inputs, but their responsiveness to distension may be augmented.

GABA(B) receptors on vagal afferent pathways: peripheral and central inhibition.

To investigate GABA(B) receptors along vagal afferent pathways, we recorded from vagal afferents, medullary neurons, and vagal efferents in ferrets. Baclofen (7-14 micromol/kg i.v.) reduced gastric tension receptor and nucleus tractus solitarii neuronal responses to gastric distension but not gastroduodenal mucosal receptor responses to cholecystokinin (CCK). GABA(B) antagonists CGP-35348 or CGP-62349 reversed effects of baclofen. Vagal efferents showed excitatory and inhibitory responses to distension and CCK. Baclofen (3 nmol i.c.v. or 7-14 micromol/kg i.v.) reduced both distension response types but reduced only inhibitory responses to CCK. CGP-35348 (100 nmol i.c.v. or 100 micromol/kg i.v.) reversed baclofen's effect on distension responses, but inhibitory responses to CCK remained attenuated. They were, however, reversed by CGP-62349 (0.4 nmol i.c.v.). In conclusion, GABA(B) receptors inhibit mechanosensitivity, not chemosensitivity, of vagal afferents peripherally. Mechanosensory input to brain stem neurons is also reduced centrally by GABA(B) receptors, but excitatory chemosensory input is unaffected. Inhibitory mechano- and chemosensory inputs to brain stem neurons (via inhibitory interneurons) are both reduced, but the pathway taken by chemosensory input involves GABA(B) receptors that are insensitive to CGP-35348.

Gregarious behavior in desert locusts is evoked by touching their back legs.

Desert locusts in the solitarious phase were repeatedly touched on various body regions to identify the site of mechanosensory input that elicits the transition to gregarious phase behavior. The phase state of individual insects was measured after a 4-h period of localized mechanical stimulation, by using a behavioral assay based on multiple logistic regression analysis. A significant switch from solitarious to gregarious behavior occurred when the outer face of a hind femur had been stimulated, but mechanical stimulation of 10 other body regions did not result in significant behavioral change. We conclude that a primary cause of the switch in behavior that seeds the formation of locust swarms is individuals regularly touching others on the hind legs within populations that have become concentrated by the environment.

Immunocytochemical localization of putative γ-aminobutyric acid receptor subunits in the head ganglia of Periplaneta americana using an anti-RDL C-terminal antibody

A polyclonal antibody raised against a 17 amino acid polypeptide (the predicted C-terminal sequence of the cloned Drosophila melanogaster γ-aminobutyric acid (GABA) receptor subunit, RDL) was used to investigate the distribution of GABA receptor subunit(s) of this type in the nervous system of the cockroach Periplaneta americana. Intense staining was detected in the calyces of the mushroom bodies, glomeruli of the antennal lobes, lower central body, the corpora cardiaca and several cell layers of the medulla and the lobula regions of the optic lobe. The most intense immunocytochemical staining was in the suboesophageal ganglion. Control sections pre-incubated with the primary antibody and conjugated peptide were not stained. Thus, it appears that a GABA receptor subunit of the RDL type is located in cockroach brain regions involved in the processing of visual, olfactory and mechanosensory inputs to the nervous system. Since the corpora cardiaca reacted to this antiserum, this type of GABA receptor may also be involved in the regulation of neurosecretory activity.

Processing of gustatory information by spiking local interneurons in the locust.

Despite the importance of gustation, little is known of the central pathways responsible for the processing and coding of different chemical stimuli. Here I have analyzed the responses of a population of spiking local interneurons, with somata at the ventral midline of the metathoracic ganglion, during stimulation of chemo- and mechanoreceptors on the legs of locusts. Volatile acidic stimuli were used to selectively activate the chemosensory neurons. Different members of the population of local interneurons received depolarizing or hyperpolarizing inputs during chemosensory stimulation. Many of the same interneurons that received chemosensory input also received mechanosensory inputs from tactile hairs on the leg, but others received exclusively mechanosensory inputs. Chemosensory inputs occurred with a short and constant latency, typical of monosynaptic connections. The chemosensory receptive fields of the spiking local interneurons mapped the surface of a hind leg so that spatial information relating to the location of a taste receptor was preserved. The amplitude of potentials in interneurons during chemosensory stimulation varied in a graded manner along the long axis of the leg, thus creating gradients in the chemosensory receptive fields of interneurons. Some interneurons were depolarized to a greater extent by chemical stimuli applied to basiconic sensilla on distal parts of the leg, whereas others were depolarized more by chemical stimulation of more proximal sensilla.

Inhibition of styloglossus motoneurons during the palatally induced jaw-closing reflex

The inhibition of hypoglossal motoneurons innervating the styloglossus muscle during transient jaw closing, the so-called jaw-closing reflex, was studied in cats. The application of diffuse pressure stimulation to the posterior palatal surface produced the jaw-closing reflex and inhibitory postsynaptic potentials in the styloglossus motoneurons, indicating that mechanosensory inputs from the posterior palatal mucosa sent inhibitory synaptic inputs to styloglossus motoneurons. We also demonstrated that, during the palatally induced jaw-closing reflex, the tongue extended at jaw closure and was still extended forward in the initial part of the opening phase. In all of 22 styloglossus motoneurons studied, the depression of firing was elicited after the onset of jaw closure. In 14 of 22 styloglossus motoneurons, the depression of firing was elicited in the closing phase, and in the remaining cells it was elicited in the occlusal phase. By increasing the intracellular concentration of chloride ions, the inhibitory postsynaptic potential elicited in the styloglossus motoneuron converted to a depolarizing potential. It is concluded that the inhibition of styloglossus motoneurons may be involved in the maintenance of tongue protrusions during the palatally induced jaw-closing reflex, and that inhibitory postsynaptic potentials evoked in the styloglossus motoneurons are partly due to a chloride-dependent inhibitory postsynaptic potential.

Differential regulation of a homologous peptidergic neuron in grasshoppers: evolution of a neural circuit

The vasopressin-like immunoreactive (VPLI) neurons of grasshoppers have paired cell bodies in the suboesophageal ganglion and both anterior and posterior running axons. In non-oedipodine grasshopper species (e.g. Schistocerca gregaria), most of their arborisations are distributed in dorsal and lateral neuropil, while in oedipodine species (e.g. Locusta migratoria), the neurons have additional extensive axonal projections in both the optic lobes and proximal portions of the ganglionic peripheral nerves. This study demonstrates that these morphological differences correlate with their physiology. In L. migratoria, VPLI neuron activity is regulated primarily via a spontaneously active interneuron which descends from the brain. This descending interneuron is inhibited by a light-activated brain extraocular photoreceptor. Regulation of VPLI neuron activity by an extraocular photoreceptor is also seen in the other oedipodine grasshopper investigated. In the four non-oedipodines examined (from two subfamilies), we find no extraocular photoreceptor regulation of VPLI neuron activity. Despite this, VPLI neuron in S.␣gregaria does appear to be driven by a descending interneuron homologous to that in L. migratoria. The descending interneuron in both species receives similar mechanosensory input and excites the VPLI neuron via cholinergic synapses. Histamine injection into the medial protocerebrum of both species causes strong inhibition of the descending interneuron. The evolution of the neural circuitry, by which an extraocular photoreceptor comes to regulate the descending interneuron in oedipodine species, is discussed.

In vitro appetitive classical conditioning of the feeding response in the pond snail Lymnaea stagnalis.

In vitro appetitive classical conditioning of the feeding response in the pond snail Lymnaea stagnalis. J. Neurophysiol. 78: 2351-2362, 1997. An in vitro preparation was developed that allowed electrophysiological analysis of appetitive conditioning of feeding in the model molluscan system, Lymnaea. The network generating the feeding motor program (fictive feeding) is well characterized at the cellular level and consists of identified central pattern generator (CPG) interneurons, motor neurons, and modulatory interneurons. Activation of a modulatory interneuron, the slow oscillator (SO), evokes the three-phase fictive feeding rhythm in the same semi-intact preparations where tactile stimuli can be applied to the lips. By pairing touch as a conditioned stimulus (CS) with stimulation of the SO as an unconditioned stimulus (US), we established an effective in vitro paradigm for appetitive conditioning. Before training, touch to the lips evoked only brief and weak activity in the feeding interneurons and motor neurons. After 6-10 conditioning trials, there was a significant enhancement in the fictive feeding response to CS alone. This was not seen in controls (CS only, US only, random CS and US) and in preparations where there was no initial brief response to touch before conditioning. Direct recordings from the protraction phase N1M interneurons during in vitro conditioning indicated that the enhancement of the fictive feeding is due to an increased activation of these CPG cells by mechanosensory inputs from the lips. We also found that the conditioned response was not due to a facilitated activation of modulatory neurons in the feeding network, such as the SO or the cerebral giant cells (CGCs), because the activity of these cells remained unchanged after conditioning.

ω‐Agatoxin IVA, a P‐Type Calcium Channel Antagonist, Reduces Nociceptive Processing in Spinal Cord Neurons with Input From the Inflamed But Not From the Normal Knee Joint — An Electrophysiological Study in the Rat In Vivo

High threshold voltage-dependent P- and Q-type calcium channels are involved in neurotransmitter release. In order to investigate the role of P- and Q-type calcium channels in the mechanosensory (nociceptive) processing in the spinal cord, their participation in the responses of spinal wide-dynamic-range neurons to innocuous and noxious mechanical stimulation of the knee and ankle joints was studied in 30 anaesthetized rats. The knee was either normal or acutely inflamed by kaolin/carrageenan. During the topical application of omega-agatoxin IVA (P-type channel antagonist, 0.1 microM) onto the dorsal surface of the spinal cord, the responses to innocuous and noxious pressure applied to the normal knee were increased to respectively 124 +/- 42% and 114 +/- 23% of predrug values (mean +/- SD, P < 0.05, 14 neurons). By contrast, in rats with an inflamed knee, the responses to innocuous and noxious pressure applied to the knee were reduced to respectively 72 +/- 19 and 73 +/- 22% of baseline (mean +/- SD, P < 0.01, 13 neurons). In the same neurons, omega-agatoxin IVA slightly increased the responses to pressure on the non-inflamed ankle whether the knee was normal or inflamed. Thus P-type calcium channels seem to acquire a predominant importance in the excitation of spinal cord neurons by mechanosensory input from inflamed tissue and hence in the generation of inflammatory pain. By contrast, the Q-type channel antagonist, omega-conotoxin MVIIC (1 or 100 microM), had no significant effect upon responses to innocuous or noxious pressure applied to either normal or inflamed knees (25 neurons).

Coordination of startle and swimming neural systems in the pteropod mollusk Clione limacina: role of the cerebral cholinergic interneuron.

The holoplanktonic pteropod mollusk Clione limacina has a unique startle system that provides a very fast, ballistic movement of the animal during escape or prey capture behaviors. The startle system consists of two groups of large pedal motoneurons that control ventral or dorsal flexions of the wings. Although startle motoneurons innervate the same musculature used during normal swimming, they are independent of the swim central pattern generator and swim motoneurons. This study demonstrates that a cerebral startle (Cr-St) interneuron, which provides prominent excitatory inputs to startle motoneurons, plays a very important role in coordination of the startle and swimming neural systems. The Cr-St interneuron produces, simultaneously with monosynaptic excitatory inputs to dorsal startle motoneurons, monosynaptic inhibitory inputs to all types of swim neurons, including interneurons of the central pattern generator, general excitor motoneurons, small motoneurons, and modulatory pedal serotonergic wing neurons. The inhibitory synaptic transmission between the Cr-St interneuron and swim interneurons and motoneurons, as well as excitatory transmission between the Cr-St interneuron and startle motoneurons, appears to be cholinergic because it is blocked by the cholinergic antagonists atropine and d-tubocurarine, mimicked by exogenous acetylcholine in very low concentrations, and enhanced by the cholinesterase inhibitor eserine (physostigmine). The Cr-St-neuron-mediated inhibitory inputs to the swimming system are strong enough to completely terminate swimming activity while the Cr-St interneuron is active. Mechanosensory inputs are capable of triggering Cr-St neuron firing at rates sufficient to suppress fictive swimming in reduced preparations. Thus the Cr-St interneuron can temporally remove the swimming system from the control over the swim musculature while simultaneously activating the startle system to produce a powerful, short-latency response.

Impact of motor activity and antennal mechanosensory input on the intensity of proctolin-like immunoreactivity in antennal motoneurons of crickets ( Gryllus bimaculatus )

The change of the intensity of proctolin immunolabeling of 2 of 17 proctolinergic antennal motoneurons [one adductor (Ad3), and one depressor (D5)] was quantitatively studied in crickets in relation to flight and antennal deafferentation. During flight, the maintained forward position of the antennae is supported by high frequency tonic firing of Ad3 but probably all motoneurons are activated. In animals sacrificed immediately after the last of five consecutive flight periods the intensity of proctolin-like immunolabeling showed a significant decrease in the Ad3 soma in comparison to the Ad3 of non-flyers. The identical result was observed in the D5 soma. In animals sacrificed 40 h after flight, no difference in the intensity of proctolin immunolabeling between the Ad3 soma of flyers and non-flyers was evident. Thus, at high motoneuron activity, the production of proctolin may not be able to keep pace with its transport from the soma. Deletion of all proprioceptors of one antenna which respond to horizontal movements only led to a significant decrease of the intensity of proctolin immunolabeling in the Ad3 soma on the operated side, but not in the soma of D5. This indicates that selected afferent input has an impact on proctolin expression in distinct motoneuron pools.

Excitation of hypoglossal motoneurons responsible for tongue protrusions is associated with palatally induced jaw-closing reflex

The excitation of hypoglossal motoneurons innervating the genioglossus and geniohyoid muscles during transient jaw closing, the so-called jaw-closing reflex, was studied in cats. The application of diffuse pressure stimulation to the posterior palatal surface produced the jaw-closing reflex, and it was found that mechanosensory inputs from the posterior palatal mucosa sent excitatory synaptic inputs to both genioglossus and geniohyoid motoneurons. We demonstrated that, during the palatally induced jaw-closing reflex, the tongue extended at jaw closure and was still extended forward in the initial part of the opening phase. In five of 27 genioglossus motoneurons and nine of 23 geniohyoid motoneurons, the onset of burst was elicited before the onset of jaw closure. The remaining cells produced the onset of burst in the closing phase and in the initial part of the occlusal phase. However, the onset of excitatory postsynaptic potentials was 75–180 ms ( n=20), earlier than that of jaw closure. During the jaw-closing reflex, the genioglossus and geniohyoid motoneurons were excited during the same phase of jaw movements and there was no difference in the onset of firing between the genioglossus and geniohyoid motoneurons. It is concluded that the excitation of the genioglossus and geniohyoid motoneurons may be associated with tongue protrusions during the palatally induced jaw-closing reflex.

The role of antennal sensory cues in female responses to courting males in the cricket Teleogryllus oceanicus

Courtship communication in the field cricket Teleogryllus oceanicus is multimodal, involving a minimum of both acoustic and chemical cues. The acoustic signal, or courtship song, is necessary to elicit normal levels of female mounting of the male during courtship. Antennal input is also crucial since antennectomized females show highly reduced levels of mounting. Immobilization of the scape&shy;pedicel and pedicel&shy;flagellum joints of female antennae had no effect on mounting probability, suggesting that mechanosensory input from chordotonal organs at the base of the antenna is not necessary. The antennal flagellum is a multimodal sensory organ, which contains both mechanoreceptors and chemoreceptors. In order to dissect the roles of different flagellar modalities, we treated antennae with zinc sulphate. This suppressed a well-characterized contact-chemosensory behaviour, initiation of courtship by males, and eliminated responsiveness of most chemosensory hairs as assayed electrophysiologically. Zinc sulphate treatment had no effect on a tactile antennal reflex, indicating that it selectively silenced chemoreceptors. Treatment of antennae with zinc sulphate reduced mounting levels nearly as much as antennectomy, suggesting that the main antennal cue required for the mounting responses is chemosensory, rather than mechanosensory, in nature.