Serotoninergic and noradrenergic projections to the ventral posterolateral nucleus of the monkey thalamus.

In the present study, serotoninergic and noradrenergic varicosities were identified in the ventral posterolateral nucleus of the macaque monkey. Monoaminergic neurons projecting to the ventral posterolateral nucleus of the thalamus were identified by using retrograde labeling with horseradish peroxidase combined with immunocytochemical staining for serotonin or dopamine-beta-hydroxylase. The midbrain nucleus raphe dorsalis was the major site of origin for neurons providing a serotoninergic projection to the ventral posterolateral nucleus. A few retrogradely labeled serotonin-containing neurons were also observed in the central superior and the raphe pontis nuclei. Noradrenergic cells with projections to the thalamus were primarily located in the nucleus locus coeruleus with some projection neurons in the nucleus subcoeruleus, and the A5 catecholamine cell group of the pons.

A method for the generation of complex vibrotactile stimuli.

A method is described utilizing computer-generated sine wave data and purpose-built hardware to generate a complex vibrotactile stimulus. Two sine waves of different frequency were summed to produce a complex waveform with two temporal components, a high frequency component and a low frequency beat component. The computer-generated data points for each of the two component sine waves were downloaded to two banks of static memory in a dual synchronous arbitrary function generator. The data points in memory were fed to two 12-bit digital-to-analogue converters which sent the two analogue sine wave signals to a summing amplifier where the two sine waves were added. This method provides a complex waveform that can be gated on and off, has a fixed frequency ratio of the component sine waves and no phase drift between the component waves. Addition of the separate sine waves in a summing amplifier allows for easy alteration of the amplitude ratio of the sine waves. The output of the summing amplifier is sent to a feedback controlled mechanical stimulator, thereby allowing the stimulus to be presented to the skin of human subjects and experimental animals.

Specificity of tactile connections in neonatal cuneate nucleus.

In the neonatal kitten, cuneate neurones activated by tactile stimulation of the forelimb foot pads are divisible into 3 functional classes each of which appears to receive its glabrous skin input selectively from a single class of tactile receptors. Their properties suggest that the functionally specific patterns of tactile convergence characteristic of the adult cuneate nucleus are established by the time of birth.

High gain transmission of single impulses through dorsal column nuclei of the cat.

Paired recordings were made in the cat from neurones of dorsal column nuclei and from intact pacinian sensory fibres of the hindlimb interosseous nerve. Direct evidence is presented for central neurones being driven by single impulses arriving over just one sensory nerve fibre. Transmission through this sensory relay appears to be optimized for the detection of minimal sensory inputs. Two mechanisms operate for the amplification of such inputs. First, individual sensory fibres can exert divergent, suprathreshold actions on multiple target neurones, and second, a single impulse coming over one input fibre can induce pairs or bursts of output spikes from its target neurones.

Representation of forelimb mechanoreceptors within the superior colliculus of the cat.

Forelimb mechanoreceptor representation within the superior colliculus was investigated using microelectrode recording in decerebrate or anaesthetized cats. Selective activation of different mechanoreceptive fibre classes supplying the distal forelimb revealed a poor representation of all three fibre classes (SA, PC and RA) that innervate the glabrous skin. Furthermore, there was no evidence for Pacinian (PC) input from the interosseous or wrist region. The tap stimuli which are often needed to activate the collicular neurones, may be effective, when applied to the distal limb, because of stimulus spread to more proximal sites, or as a result of the synchronous recruitment of multiple classes of afferent fibres.

A simple, accurate method for reproducing details of histological sections using a microfiche printer.

A rapid, inexpensive method is described for achieving accurate reproduction of histological sections. The method uses a microfiche reader-printer (for library use) which produces A4 size prints. Interchangeable lenses on the microfiche printer permit the magnification of the section to be varied over the range 6.6-72 times. Where large numbers of sections are involved the speed and low cost of the method offers considerable advantages over traditional hand tracing or photomicrography.

Functional properties of slowly adapting mechanoreceptors in cat footpad skin.

The functional properties of slowly adapting (SA) afferent fibers innervating cat footpad skin were examined. Measurements were taken of receptive field area; spontaneous activity (less than 1 impulse/sec); the slope of the stimulus-response curve for steady indentations up to 2 mm in amplitude; variability of the interimpulse intervals, as measured by the coefficient of variation of time interval histograms; decay of the response to steady indentation; and sensitivity to sinusoidal vibration (most sensitive at 5-10 Hz). Where comparable tests were performed on glabrous and hairy skin SA fibers, the functional properties of those in glabrous skin more closely resembled SAI fibers than SAII fibers. Additional results from glabrous skin SA fibers suggest that it is distortion of the nerve endings rather than steady indentation or compression that leads to a brisk response. On the measures described above, there appeared to be only one functional class of SA fiber innervating the cat footpad skin.

Functional capacities of tactile afferent fibres in neonatal kittens.

1. Responses were recorded from individual tactile afferent fibres isolated by microdissection from the median nerve of pentobarbitone-anaesthetized neonatal kittens (1-5 days post-natal age). Experiments were also conducted on adult cats to permit precise comparisons between neonatal and adult fibres.2. Neonatal fibres with receptive fields on the glabrous skin of the foot pads were classified into two broad groups, a slowly adapting class (40%) which responded throughout a 1 sec period of steady indentation and a rapidly adapting or dynamically sensitive class comprising 60% of units. Fibres in these two groups had overlapping conduction velocities in the range 4.3 to 7.5 m/sec and were believed to be the developing Group II afferents of the adult.3. Neonatal slowly adapting fibres qualitatively resembled their adult counter-parts. They displayed graded stimulus-response relations which, over the steepest segment of the curves, had mean slopes of 15.7 impulses/100 mum of indentation. Plateau levels of response were often reached at amplitudes of skin indentation of < 0.5-0.7 mm.4. Dynamically sensitive fibres with receptive fields on the glabrous skin were studied using sinusoidal cutaneous vibration which in the adult enables them to be divided into two distinct classes. However, in the neonate, they formed a continuum whether criteria of sensitivity or responsiveness were used.5. In response to vibration neonatal fibres differed from adult ones according to the following quantitative indices: (i) sensitivity as measured by both absolute thresholds and thresholds for a 1: 1 pattern of response, both of which were higher in the neonate than in the adult at all frequencies > 50 Hz and differed by an order of magnitude at frequencies >/= 200 Hz; (ii) responsiveness based on the mean impulse rate evoked at a fixed amplitude of cutaneous vibration; (iii) band width of vibratory sensitivity which in the neonate was confined to approximately 5-300 Hz whereas in the two classes of adult units it covered the range 5-800 Hz; (iv) capacity for coding information about vibration frequency. Impulse activity of neonatal fibres was less tightly phase-locked to the vibratory stimulus and showed a poorer reflection of the periodic nature of the vibratory stimulus than impulse patterns of adult units.6. The results reveal that tactile receptors and afferent fibres in the neonate are functionally immature. Their restricted coding capacities suggest that peripheral tactile sensory mechanisms impose limits on the ability of the new-born animal to derive information about its tactile environment.

Responses of spinothalamic tract cells in the superficial dorsal horn of the primate lumbar spinal cord.

1. The responses of thirty-five spinothalamic tract (s.t.t.) cells in or near lamina I of the dorsal horn were examined in chloralose- and barbiturate-anaesthetized monkeys (Macaca fascicularis). Many of the cells could be classified on the basis of receptive field properties as either wide dynamic range (w.d.r.) cells or as high-threshold (h.t.) cells. 2. Thalamic stimulation sites for antidromic activation of the s.t.t. cells were in or around the ventral posterior lateral nucleus. Axons of the s.t.t. cells had a mean conduction velocity of 17 m/s (33 and 14 m/s for w.d.r. and h.t. cells, respectively). Mean minimum afferent conduction velocity averaged 37 m/s (52 and 23 m/s for w.d.r. and h.t. cells, respectively). Background activity was low (mean of 2.3 impulses/s). 3. An alternative classification of the cells was based on a kappa means cluster analysis of the responses to a series of mechanical stimuli. The response profiles for a given cell were normalized, and those of the s.t.t. cells in or near lamina I were analysed along with the responses of a population of s.t.t. cells, largely in laminae IV-VI, that had been described previously. S.t.t. cells in or near lamina I were distributed amongst three of the four groups of cells determined by the cluster analysis (types 2-4). 4. Vibratory stimuli excited most of the w.d.r. but none of the h.t. cells tested. Best frequencies were 5-10 Hz (at 100 and 500 microns indentations). 5. Most w.d.r. but few h.t. cells responded to cutaneous cooling. All of the cells responded to noxious heating, but w.d.r. cells had steeper stimulus-response curves. 6. After a series of noxious heat stimuli, the thresholds for noxious heat were lowered and responses to lower-intensity noxious heat stimuli were enhanced (sensitization). However, responses to more intense stimuli were reduced (inactivation). Similar changes were seen in the responses to graded mechanical stimuli. 7. It is concluded that s.t.t. cells in or near lamina I can signal noxious cutaneous stimuli but have poor coding abilities for innocuous mechanical stimuli. Some of these cells respond to innocuous thermal stimuli, but their role in thermoreception is unclear. The small receptive fields suggest that these cells could contribute to stimulus localization.

Temporal patterning in the responses of gracile and cuneate neurones in the cat to cutaneous vibration.

1. Recordings were made in decerebrate cats from gracile and cuneate neurones responding to vibration-induced inputs from Pacinian corpuscle (P.c.) receptors of the hind-limb and forelimb footpads. The two groups of neurones were compared, in particular for their capacities for responding to cutaneous vibration with phase-locked impulse patterns. 2. In both nuclei the P.c. neurones were most sensitive to vibration in the range 80 to greater than 600 Hz. Stimulus-response relations were similar for the two groups, as were measures derived from these relations such as response levels, absolute thresholds and the dynamic range (defined as the vibration amplitude range over which responses were graded). 3. At frequencies up to 300-400 Hz, responses for some neurones in both nuclei remained well phase locked to the vibration; however, quantitative analysis using a factorial analysis of variance indicated that the phase locking was poorer in gracile than cuneate neurones. 4. In both nuclei there was marked variability from neurone to neurone in measures of phase locking which may reflect variations in the extent of convergence of P.c. fibres upon different target neurones. For neurones in either nucleus that had comparatively tight phase locking of responses to vibration it is proposed that their output is functionally dominated by one or a few of their convergent P.c. input fibres.

Actions of single sensory fibres on cat dorsal column nuclei neurones: vibratory signalling in a one-to-one linkage.

1. The synaptic linkage between single, identified sensory fibres associated with Pacinian corpuscle (P.c.) receptors and central neurones of the dorsal column nuclei was examined in decerebrate or anaesthetized cats. Paired recordings were made from individual neurones in the gracile division of the dorsal column nuclei and from the hind-limb interosseous nerve in which it is possible to identify and monitor the activity of each P.c. fibre activated when recording from the intact nerve with a platinum hook electrode. Individual P.c. fibres were activated by vibration delivered with an 0.2 mm diameter probe to the interosseous P.c. receptors. 2. Thirty-five P.c. fibre-gracile neurone pairs were isolated in which activity in the single, identified P.c. fibre evoked suprathreshold responses (mean latency +/- S.D., 10.3 +/- 1.5 ms) in the gracile neurone. A single impulse arriving over one P.c. fibre could generate pairs or triplets of output spikes from several target neurones thus revealing a potent synaptic organization within the dorsal column nuclei for the transmission and amplification of weak sensory signals. 3. The potency of the linkage for some pairs resulted in post-synaptic response levels of up to 400 impulses s-1 when a single input fibre was discharging one impulse on each vibration cycle at 200-400 Hz. 4. Gracile neurones driven by single P.c. fibres had phase-locked responses to vibration at frequencies of up to 400-500 Hz. However, the responses displayed much greater phase dispersion than those of P.c. fibres, indicating that a major component of phase dispersion in the vibration-induced responses of dorsal column nuclei neurones is attributable to the properties of the synaptic linkage between an individual fibre and the target neurone. 5. The potent actions of single, identified P.c. fibres on their target neurones are consistent with the hypothesis that phase-locked responses in dorsal column nuclei neurones to vibration at 100-400 Hz may reflect the functional domination of the target neurone's output by one or a few of its converging fibres.

Integrative processing of vibratory information in cat dorsal column nuclei neurones driven by identified sensory fibres.

1. In decerebrate or anaesthetized cats, the vibration-induced responses of dorsal column nuclei neurones were examined, first, when their input came from simultaneously recorded pairs or other combinations of identified Pacinian corpuscle (P.c.) afferent fibres of the interosseous nerve, and secondly, when different convergent sets of P.c. fibres were engaged by footpad vibration. 2. Suprathreshold actions were observed on individual dorsal column nuclei neurones from two or more identified P.c. fibres. Recruitment of these convergent fibres usually led to summation in the dorsal column nuclei neurone as reflected in higher response levels compared with those evoked by single-fibre inputs. 3. When the input was increased from one to two or more identified P.c. fibres the dorsal column nuclei neurones could retain a single, dominant phase of response to high-frequency (greater than 100 Hz) vibration even though these fibres, in isolation, evoked responses in the target neurone at substantially different latencies. However, on average, phase locking was significantly tighter in response to single-fibre input than to multiple P.c.-fibre input. 4. Dorsal column nuclei neurones were also able to retain phase-locked responses to high-frequency vibration when phase differences between different convergent inputs were systematically introduced to alter the degree of synchrony in the activity arriving over convergent, identified P.c. fibres. 5. When the input to dorsal column nuclei neurones came from the skin it was found that with the recruitment of two converging sets of P.c. fibres the dorsal column nuclei neurones were able to retain phase-locked responses to high-frequency vibration even when phase shifts were introduced between the two sets of P.c. inputs. 6. In conclusion, the observed integrative processing by dorsal column nuclei neurones of vibration-induced inputs arriving over identified, convergent P.c. fibres, or sets of P.c. fibres, is consistent with our hypothesis that the retention of phase-locked responses to vibration at frequencies greater than or equal to 100 Hz may reflect the functional domination of the target neurone by just one or a few of its convergent input fibres.

Functional maturation of tactile sensory fibers in the kitten.

The maturation of tactile coding capacities was investigated in sensory fibers supplying the forelimb footpads in anesthetized kittens. Fibers were isolated by microdissection from the median or ulnar nerves of neonatal kittens (1st to 5th postnatal day) and kittens in the age categories 10-15, 25-30, 55-63, and 83-90 days. The use of quantitative, reproducible tactile stimuli, in particular, cutaneous vibration, and objective analytical procedures enabled response parameters to be quantified and compared at different ages with those of adult fibers. While three classes of myelinated tactile sensory fibers are associated with the footpads in adult cats, one of them slowly adapting (SA) and two, the rapidly adapting (RA) and pacinian corpuscle (PC) classes, showing pure dynamic sensitivity, this breakdown into three classes is not apparent until 10-15 days after birth. In all age groups, the SA fibers displayed responses that were graded depending on the magnitude of the skin indentation. However, in younger kittens (less than 25-30 days) plateau levels of response were sometimes attained over a narrow range (0.5 mm) of indentation. From 10-15 days, when RA fibers are identifiable, their sensitivity, as measured by absolute and 1:1 thresholds to cutaneous vibration, was independent of age at low vibration frequencies, less than or equal to 80 Hz. However, at high frequencies, e.g., 200 Hz, their thresholds appeared to decrease during the 1st postnatal month with a resulting expansion in their bandwidth of vibratory sensitivity. At their best frequency, around 30 Hz, RA fibers appear mature by 10-15 days in their capacity for encoding vibratory frequency information. The PC class of tactile afferents displayed the most striking functional changes with age. Their vibratory bandwidths, at 50-micron amplitude, expand from an upper limit of about 200-300 Hz in the neonate to mature values of about 800-1,000 Hz over a 2-mo postnatal period.(ABSTRACT TRUNCATED AT 400 WORDS)

Tactile sensory coding during development: signaling capacities of neurons in kitten dorsal column nuclei.

The functional maturation of cuneate neurons was studied using reproducible tactile stimulation procedures and quantitative assessment of coding capacities in anesthetized (N2O/O2 plus barbiturate infusion) or decerebrate cats from six different age groups; neonatal (1-5 days), 10-15, 25-30, 55-63, and 83-90 days, and an adult group. Cuneate neurons were distinguished from input fibers to the nucleus on criteria of spike configuration and time course and on response profiles. Extracellular spike durations underwent a progressive shortening with age, reaching the adult range at approximately 3 mo when background activity levels also reached maturity. Despite an increase in conduction-path length, response latencies decreased to adult values by 25-30 days of age, presumably reflecting the increased conduction velocity in the input fibers. In each age group three functional classes of neurons responsive to tactile stimulation of the footpads were identifiable. One consisted of slowly adapting neurons whose stimulus-response relations resembled those of the adult in responsiveness and dynamic range by 25-30 days postnatal age. The remaining neurons were purely dynamically sensitive, but among them two classes could be distinguished by their differential sensitivity to cutaneous vibration, one receiving rapidly adapting (RA) fiber input the other pacinian corpuscle (PC) input. The combined bandwidth of vibration sensitivity in dynamically sensitive cuneate neurons expands from approximately 5-300 Hz in the neonate to the mature range of 5-1,000 Hz by 1 mo of age. The PC class of cuneate neurons showed an upward shift in peak sensitivity from 30-200 Hz in the neonate to 100-300 Hz at 10-15 days, reaching adult values of 200-600 Hz by 25-30 days. Over this period absolute thresholds dropped by an order of magnitude at 200-500 Hz reflecting the threshold trends in primary PC fibers. During the first 1-2 postnatal months, the capacity of cuneate PC neurons to signal information in a pattern code about vibration frequencies around 300 Hz is restricted because of poor phase locking and low responsiveness, which preclude an impulse periodicity reflecting the vibratory frequency.(ABSTRACT TRUNCATED AT 400 WORDS)

Primate nucleus gracilis neurons: responses to innocuous and noxious stimuli.

1. Recordings were made from 67 neurons in the nucleus gracilis (NG) of anesthetized macaque monkeys. All of the cells were activated antidromically from the ventral posterior lateral (VPL) nucleus of the contralateral thalamus. Stimuli used to activate the cells orthodromically were graded innocuous and noxious mechanical stimuli, including sinusoidal vibration and thermal pulses. 2. The latencies of antidromic action potentials following stimulation in the VPL nucleus were significantly shorter for cells in the caudal compared with the rostral NG. The mean minimum afferent conduction velocity of the afferent conduction velocity of the afferent fibers exciting the NG cells was 52 m/s, as judged from the latencies of the cells to orthodromic volleys evoked by electrical stimulation of peripheral nerves. The overall conduction velocity of the pathway from peripheral nerve to thalamus was approximately 40 m/s. 3. Cutaneous receptive fields on the distal hindlimb usually occupied an area equivalent to much less than a single digit. However, a few cells had receptive fields up to or exceeding the area of the foot. 4. NG cells were classified by their responses to graded mechanical stimulation of the skin as low threshold (LT) or wide dynamic range (WDR). No high-threshold NG cells were found. A special subcategory of pressure-sensitive LT (SA) neurons was recognized. Many of these cells were maximally responsive to maintained indentation of the skin. The sample of NG cells differed from the population of primate spinothalamic and spinocervicothalamic pathways so far examined, in having a larger proportion of LT neurons and a smaller proportion of WDR cells. A few NG cells responded best to manipulation of subcutaneous tissue. 5. Discriminant analysis permitted the NG cells to be assigned to classes determined by a k-means cluster analysis of the responses of a reference set of 318 primate spinothalamic tract (STT) cells. There were four classes of cells based on normalized responses of individual neurons and another four classes based upon responses compared across the population of cells. The NG cells were allocated to the various categories in different proportions than either primate STT cells or spinocervicothalamic neurons, consistent with the view that the functional roles of these somatosensory pathways differ. 6. Some of the pressure-sensitive NG cells were excited when the skin was stretched, suggesting an input from type II slowly adapting (Ruffini) mechanoreceptors.(ABSTRACT TRUNCATED AT 400 WORDS)

Somatotopic organization and response characteristics of dorsal horn neurons in the cervical spinal cord of the cat.

Somatotopic organization was examined for 203 dorsal horn cells in spinal segments C6 and C7 of chloralose-anesthetized cats. The ventral paw and toe area were represented medial to a smaller area with input from the dorsal paw. Representation of the ventromedial forelimb was rostral to that of the paw, while the shoulder and dorsolateral limb were represented caudal to it. In 13 out of 22 electrode tracks in which three or more cells were found, the location of receptive fields progressively changed for successively recorded cells. Receptive fields on the paw were closer together and overlapped more than those on the proximal limb. Receptive fields that included glabrous skin were found for only 7 of 203 cells; all were located in the medial third of the C7 dorsal horn. It appears that glabrous skin is underrepresented in the dorsal horn; this may be compensated for a higher levels by input from the lemniscal system. The response characteristics of 172 dorsal horn neurons were examined. Of these units, 135 (78%) had cutaneous receptive fields. An additional 37 cells (22%) responded to manipulation of muscle or tendon and were classified as deep (D) cells. The cells with cutaneous receptive fields were classified as low-threshold (LT) cells (38%), high-threshold (HT) cells (20%), and wide-dynamic-range (WDR) cells (20%). Alternatively, using cluster analysis, 57 cells with cutaneous receptive fields were classified as one of five mechanical types. Type 1 cells responded primarily to low-threshold input, while the other four types fired in characteristic patterns in response to a combination of innocuous and noxious stimuli. LT cells were located more superficially in the spinal cord than the other classes; their average depth below the cord surface was 1.9 mm. WDR cells (mean = 2.1 mm) were located below the LT cells and above the HT and D cells (mean = 2.6 mm).

Responses of spinothalamic tract cells in the cat cervical spinal cord to innocuous and graded noxious stimuli.

The response properties of spinothalamic tract (STT) cells in the dorsal horn of the cervical spinal cord were examined in chloralose-anesthetized cats. The activity of 56 STT cells located in laminae IV-VI was studied, with most activity isolated in the lateral part of the dorsal horn. The level of background activity in STT cells was low (mean = 1.2 impulses/sec; n = 26). Conduction velocity estimates for STT axons ranged from 9 to 76 m/sec (mean = 38 m/sec; n = 56) and were not correlated with the recording site in the spinal cord. Most cells were antidromically activated from an electrode in the medial part of the posterior group of nuclei in the thalamus. Excitatory receptive fields were ipsilateral to the recording site, and for 38 of 40 neurons were confined to the forelimb. Although receptive fields were often restricted to part of the paw, they did not include glabrous skin. Among 31 cells classified, four groups were identified: low-threshold (LT) cells (13%) responded to pressure and brushing of the skin; high-threshold (HT) cells (13%) responded only to noxious pinching or squeezing of the skin; wide-dynamic-range (WDR) cells (58%) responded to innocuous mechanical stimuli but had a greater response to noxious stimuli; deep (D) cells (16%) responded to manipulation of subcutaneous tissues such as muscle. Heat stimuli 30 sec in duration, in the range of 43-55 degrees C, were applied to the receptive fields of 14 neurons that included representatives from all three groups with cutaneous input. Nine neurons responded to heat with thresholds that ranged from 47 degrees to 55 degrees C (mean = 51 degrees C). The responses of these nine STT cells increased with increasing stimulus intensity in the noxious range. In the cat cervical dorsal horn, STT cells can signal the occurrence of noxious stimuli on the body surface, and, judging by the sizes of their peripheral receptive fields, are capable of signaling precise information about the location of the damage. Furthermore, some cells are able to signal the intensity of a noxious heating pulse.

Coding of information about tactile stimuli by neurones of the cuneate nucleus.

1. The responses of cuneate neurones to controlled tactile stimulation of the foot pads were examined in unanaesthetized, decerebrate cats. The neurones were divided into three functional classes; one sensitive to steady tactile stimuli, and two dynamically sensitive classes which could be readily differentiated by their responsiveness to cutaneous vibration. Each class appeared to receive an exclusive input from only one of the three known groups of tactile receptors associated with the foot pads, namely the Pacinian corpuscles, the Merkel endings and the intradermal, encapsulated endings known as Krause or Meissner corpuscles. 2. Cuneate neurones responsive to steady indentation of the skin displayed approximately linear or sigmoidal stimulus-response relations over indentation ranges up to approximately 1.5--2 mm. Response variability at a fixed stimulus intensity was relatively low and showed little systematic change over the full range of the stimulus-response curves. 3. One class of dynamically sensitive cuneate neurones responded to cutaneous vibration over a range of approximately 5-80 Hz with maximal responsiveness around 30 Hz. The other class, the Pacinian neurones, responded over a range of approximately 80- greater than 600 Hz with maximal responsiveness at 200-400 Hz. The thresholds and combined band width of vibratory sensitivity of these populations were comparable with known subjective thresholds and range of cutaneous vibratory sensibility. 4. Responses of cuneate neurones were phase-locked to the vibratory stimulus suggesting that information about vibration frequency could be coded by the patterns of impulse activity. Quantitative measures indicated that maximal phase-locking occurred in responses to vibration frequencies of 10-50 Hz with a progressive decline at higher frequencies. Above 400 Hz, impulse activity occurred almost randomly throughout the vibratory stimulus cycle and therefore carried little further signal of vibratory frequency. The decline, with increasing frequency, in the ability of cuneate neurones to signal information about vibratory frequency parallels the known subjective capacities for frequency discrimination. 5. A switch-over occurred at approximately 80 Hz in the population of cuneate neurones able to provide the more reliable signal of vibratory frequency; above 80 Hz, the Pacinian neurones; below 80 Hz, the neurones receiving intradermal, rapidly adapting receptor input from the pads. 6. The observed properties of cuneate neurones are compatible with a role in signalling information which could contribute to subjective tactile abilities.

Human tactile detection thresholds: modification by inputs from specific tactile receptor classes.

1. Human detection thresholds for a vibratory stimulus applied to the volar surface of the index finger were examined under conditions where afferents from specific tactile receptor classes were simultaneously activated from the thenar eminence. The experiments were designed to test whether stimuli which have been shown previously to induce afferent inhibition of ;tactile' neurones in the cuneate nucleus of the cat could modify human subjective performance in a tactile detection task. Conditioning stimuli to the thenar eminence were usually of three forms; steady indentation to engage slowly adapting tactile receptors; 300 Hz vibration to engage Pacinian corpuscles; and 30 Hz vibration to engage the intradermal, rapidly adapting tactile receptors which are thought to be Meissner's corpuscles.2. In ten subjects the mean detection threshold for a 30 Hz test stimulus in the absence of conditioning stimulation was 8.6 +/- 1.0 mum (S.E.). Detection thresholds were increased substantially in the presence of a 300 Hz, 100 mum conditioning stimulus (mean increase 11.1 +/- 2.0 mum), whereas minor or insignificant effects were seen with conditioning stimuli consisting of (a) 30 Hz, 100 mum (mean increase 1.4 +/- 0.8 mum), (b) steady indentation, 1.5 mm in amplitude (mean increase 1.3 +/- 0.7 mum) or (c) 300 Hz, 100 mum to the contralateral thenar eminence (mean increase 0.4 +/- 0.5 mum).3. The 300 Hz conditioning stimulus to the ipsilateral thenar eminence caused a marked increase in detection thresholds at all test stimulus frequencies over the range 10-450 Hz. The effects of the conditioning stimulation therefore operated on inputs from Pacinian corpuscles, which are responsible for vibration detection at 80-450 Hz, and on inputs from the intradermal, rapidly adapting receptors which are responsible for vibration detection at 10-80 Hz.4. The band width of conditioning vibratory frequencies which was effective at amplitudes of 100 mum in bringing about increases in detection threshold extended from 50-80 Hz to 300 Hz, the maximum tested.5. Whereas amplitudes of 1-2 mum produced clear increases in detection thresholds with conditioning stimuli of 300 Hz, amplitudes of > 200 mum were needed at 30 Hz.6. The observed elevations in detection threshold are consistent with an afferent-induced inhibitory action exerted at synaptic relays of the sensory pathway by tactile inputs arising exclusively or predominantly from Pacinian corpuscles.

The primate spinocervicothalamic pathway: responses of cells of the lateral cervical nucleus and spinocervical tract to innocuous and noxious stimuli.

1. The response properties of neurons of the spinocervicothalamic pathway were studied in anesthetized macaque monkeys. Graded innocuous and noxious mechanical stimuli, including sinusoidal vibration and thermal pulses, were applied to the cutaneous receptive fields. 2. Forty-nine cells in the lateral cervical nucleus (LCN) were identified by antidromic activation from the ventral posterior lateral (VPL) nucleus of the contralateral thalamus. Twelve spinocervical tract (SCT) cells in the lumbosacral enlargement of the spinal cord were identified by antidromic activation from stimulation of the ipsilateral dorsolateral funiculus below C3 but not above C1. 3. Latencies for antidromic activation of LCN neurons averaged 2.3 ms, corresponding to a mean conduction velocity of approximately 17 m/s. Mean latency for orthodromic activation of LCN neurons following electrical stimulation of peripheral nerves was 12.6 ms. Overall mean conduction velocity for the monkey spinocervicothalamic pathway was estimated to be 29 m/s. 4. Most LCN cells had receptive fields on hairy skin, but some had input from glabrous skin and a few had subcutaneous fields. The receptive fields of most SCT cells had a glabrous skin component. Receptive fields tended to be smaller for SCT than LCN cells even for fields on a comparable part of the distal hindlimb. 5. Based on their responses to a series of mechanical stimuli (brushing, pressure, pinch, and squeeze), LCN and SCT cells were classified as low-threshold (LT), wide dynamic range (WDR), or high-threshold (HT) neurons. Most of the cells were in the LT or WDR classes. Thus the spinocervicothalamic pathway in the monkey differs from the spinothalamic tract (STT), in that STT cells are generally of the WDR or HT classes. 6. With the use of discriminant analysis, LCN and SCT neurons were allocated to categories determined from a k-means cluster analysis of the responses of 318 STT cells. The LCN and SCT neurons were in different proportions in the various categories than were STT cells, suggesting differences in the signaling properties of the spinocervicothalamic and spinothalamic paths. 7. Innocuous steady indentation of the skin failed to excite any of the neurons tested. Thus no positive evidence was obtained for an input to LCN neurons from slowly adapting mechanoreceptors. 8. Sinusoidal vibratory stimuli were used to test the ability of LCN and SCT neurons to follow repeated innocuous mechanical stimuli. Vibration at 10 Hz and an amplitude of 100 micron resulted in repetitive discharges in most LCN neurons and half the SCT neurons tested; many LCN neurons had thresholds below 25 micron.(ABSTRACT TRUNCATED AT 400 WORDS)