Oculomotor areas of the primate frontal lobes: a transneuronal transfer of rabies virus and [14C]-2-deoxyglucose functional imaging study.

We used the [14C]-2-deoxyglucose method to study the location and extent of primate frontal lobe areas activated for saccades and fixation and the retrograde transneuronal transfer of rabies virus to determine whether these regions are oligosynaptically connected with extraocular motoneurons. Fixation-related increases of local cerebral glucose utilization (LCGU) values were found around the fundus of the inferior limb of the arcuate sulcus (AS) just ventral to its genu, in the dorsomedial frontal cortex (DMFC), cingulate cortex, and orbitofrontal cortex. Significant increases of LCGU values were found in and around both banks of the AS, DMFC, and caudal principal, cingulate, and orbitofrontal cortices of monkeys executing visually guided saccades. All of these areas are oligosynaptically connected to extraocular motoneurons, as shown by the presence of retrogradely transneuronally labeled cells after injection of rabies virus in the lateral rectus muscle. Our data demonstrate that the arcuate oculomotor cortex occupies a region considerably larger than the classic, electrical stimulation-defined, frontal eye field. Besides a large part of the anterior bank of the AS, it includes the caudal prearcuate convexity and part of the premotor cortex in the posterior bank of the AS. They also demonstrate that the oculomotor DMFC occupies a small area straddling the ridge of the brain medial to the superior ramus of the AS. Our results support the notion that a network of several interconnected frontal lobe regions is activated during rapid, visually guided eye movements and that their output is conveyed in parallel to subcortical structures projecting to extraocular motoneurons.

Tuning in caudal fastigial nucleus units during natural and galvanic labyrinth stimulation.

Neurons of the caudal fastigial nucleus were investigated by means of single unit recordings. Natural vestibular stimuli were applied as well as galvanic labyrinth polarization. One-third of the neurons showed a convergence of vertical and horizontal canals. More than 80% of the neurons responded to polarization of both the ipsilateral and contralateral canals (binaural responders). Most neurons had a limited response range. Two classes of neurons could be distinguished: up to 1 Hz responders and up to 10 Hz responders. In addition a group of fastigial cells showed a tuning within a small range of frequencies (sharp-tuning responders).

Corticofugal connections between the cerebral cortex and the vestibular nuclei in the rat.

The distribution of cortical efferent connections to the vestibular nuclei was quantitatively analyzed by means of retrograde axonal transport of horseradish peroxidase, wheat germ agglutinin-horseradish peroxidase, and Fast Blue in rats. The tracer substances were injected into the spinal vestibular nucleus (SpVe), the caudal part of the medial vestibular nucleus (MVe), and nucleus X of Brown Norwegian rats. Projections to the vestibular nuclei were revealed bilaterally, but predominantly contralaterally from five cortical areas: (1) the parietotemporal region (PT) which occupied the caudal two-thirds of the secondary somatosensory area and spread over the caudal part of the primary somatosensory area and the visceral cortex; (2) the anterior forelimb (AF) overlapping the anterior part of the forelimb area and the transitional zone; (3) the anterior hindlimb (AH) overlapping the anterior part of the hindlimb area and the transitional zone; (4) the lateral forelimb (LF) centered in the intercalated zone lateral to the forelimb area; and (5) the ventrotemporal region (VT) located at the ventral part of the temporal cortex. In addition to these cortical fields, the frontal cortex was found to project directly to the vestibular nuclei. These corticofugal projections were verified in experiments in which biocytin was injected into the rat PT. Anterogradely labelled fibers were traced predominantly contralaterally to the SpVe, caudal part of the MVe, and nucleus X. It is suggested that the rat corticofugal projections to the caudal vestibular nuclei modify vestibular reflexes to assist in coordinating eye, head and body movements during locomotion.

Variable otolith contribution to the galvanically induced vestibulo-ocular reflex.

The torsional eye movements elicited by sinusoidal galvanic vestibular stimulation (GVS) (0.012-3.13 Hz) were examined in healthy humans. GVS consistently induced sinusoidal modulation of the torsional slow phase velocity (SPV), which was linearly related to stimulus intensity. At low frequencies (< 0.1 Hz) nystagmic responses could be discriminated from an underlying 'tonic' modulation of eye position, which was prominent in some, but negligible in other subjects, and was not correlated with the SPV modulation. The actual SPV modulation consistently exceeded the (hypothetical) velocity modulation derived from the tonic positional components, albeit variably by almost 20-fold across subjects. This indicates that the contribution of possibly otolith-related response components to the galvanic vestibulo-ocular reflex may vary considerably in normal individuals.

Co-localization of glutamate, choline acetyltransferase and glycine in the mammalian vestibular ganglion and periphery.

Glutamate (Glu) is considered to be the main transmitter at the central synapses of primary vestibular afferents (PVA) and glycine (Gly) is assumed to play a modulatory role. In the vestibular periphery a transmitter role for acetylcholine (ACh) has been attributed chiefly to vestibular efferents (VE), however only a subset of VE neurons displays immunoreactivity (ir) for choline acetyltransferase (ChAT) and acetylcholine esterase (AChE). Controversial results exist on the presence of these two enzymes in PVA. In this study the presence of Glu, ChAT, Gly and their co-localization in the vestibular ganglia (VG) and end organs of mouse, rat, guinea pig and squirrel monkey were investigated. In the VG all bipolar neurons display strong Glu-ir and the majority of cells show a graded ChAT-ir and Gly-ir in all species examined. ChAT and Gly are present in highly overlapping neuronal populations and with a similar gradation. In the end organs ChAT and Gly are again co-localized in the same sets of fibers and endings. In conclusion, in the vestibular ganglion and end organs ChAT appears also to be present in primary afferents rather than being restricted to efferent processes. ChAT in primary afferents might indicate a modulatory or co-transmitter function of acetylcholine.

Unbiased number of vestibular ganglion neurons in the mouse.

Vestibular sensory information from the labyrinth and otolith organs is conducted to the central nervous system exclusively via primary vestibular afferents (PVA) originating from neurons located in the vestibular ganglion (VG). In the present study, the total number of VG neurons was determined in two different wild-type mouse strains using the principles of unbiased stereological counting methods by means of the physical disector. 3316 (+/-225 SD) neurons were present in the VG of the B6CBA-strain and 3551 (+/-239 SD) in C57BL/6J-mice. Since no statistical difference was detected between the two strains, the pooled mean number was 3433 (+/-232 SD) neurons. This is the first unbiased estimate of VG neurons aimed at providing a numerical basis for comparative studies and for the impact of experimental, pharmacological and pathological conditions as well as ageing on the survival and maintenance of VG neurons.

Is there a vestibular cortex?

Very different areas of the primate cortex have been labelled as 'vestibular'. However, no clear concept has emerged as to where and how the vestibular information is processed in the cerebral cortex. On the basis of data from single-unit recordings and tracer studies, the present article gives statistical evidence of the existence of a well-defined vestibular cortical system. Because the data presented here have been verified in three different primate species, it can be predicted that a similar vestibular cortical system also exists in humans.

Vestibular ganglion neurons survive the loss of their cerebellar targets.

Neuronal survival during mammalian development crucially depends on target-derived neurotrophic factors. Target loss removes this trophic support and leads in most cases to the transsynaptic retrograde degeneration of the respective afferents. Primary vestibular afferents (PVA) originating from bipolar neurons in the vestibular ganglion (VG) are the first mossy fibers that enter the cerebellum, but little is known about the survival requirements of VG neurons. In the present study the influence of the differential granule cell (GC) target loss on the survival of VG neurons was studied quantitatively using unbiased stereological methods in the cerebellar mutants Purkinje cell degeneration (pcd/pcd), Lurcher (Lc/+), and Weaver (wv/wv). Neither the secondary GC loss in the Purkinje cell deficient mutants pcd/pcd and Lc/+, nor the primary loss of GCs in wv/wv produced any significant reduction in the total number of bipolar neurons in the VG compared to controls. So, PVA neurons are highly resistant to cerebellar target deprivation and survive in the absence of cerebellar granule and Purkinje cells, regardless of whether the target loss occurs before (in wv/wv), during (in Lc/+) or after (in pcd/pcd) the mossy fiber-granule cell synaptogenesis.

Co-localization of glycine and calbindin D-28k in the vestibular ganglion of the rat.

Bipolar neurons of the vestibular ganglion (VG) are biochemically heterogeneous. The calcium-binding protein calbindin D-28k (Calb) is present only in a subset of particularly large neurons, and the amino acid glycine (Gly) has been immunocytochemically detected in a group of similarly sized cells. The close correspondence in size and number of cells in these two subgroups suggests that the Calb- and Gly-positive populations may be identical. In order to test this hypothesis, we performed direct and indirect double-labeling for Calb and Gly in the VG of the rat. The results confirm the existence of a distinct subpopulation of Calb-immunoreactive neurons, consisting of the largest cells in the VG. In contrast, the vast majority of neurons in the VG display some degree of Gly immunoreactivity, which gradually decreases from intense to almost unlabeled. Direct evidence is provided that the fraction of cells most heavily labeled by Gly antibodies is not identical with the Calb-positive subpopulation. Although some correlation between soma diameter and labeling intensity exists, Gly immunoreactivity is clearly not restricted to large neurons. The findings imply that the functional mechanisms in which Gly is potentially involved may be shared by a large spectrum of primary vestibular afferents with a broad range of physiological properties.

Corticofugal connections between the cerebral cortex and brainstem vestibular nuclei in the macaque monkey.

The distribution of cortical efferent connections to brainstem vestibular nuclei was quantitatively analysed by means of retrograde tracer substances injected into different electrophysiologically identified parts of the brainstem vestibular nuclear complex of five Java monkeys (Macaca fascicularis). Three polysensory vestibular areas were found to have a substantial projection to the vestibular nuclei: area 2v located at the tip of the intraparietal sulcus, the parietoinsular vestibular cortex (PIVC) covering the most occipital part of the granular insula (Ig) and the retroinsular area (Ri or reipt), and the dorsolateral part of the somatosensory area 3a ("area 3aV" neck/trunk region). From physiological recording experiments, these three cortical fields were known to contain many neurons responding to stimulation of semicircular canals as well as to optokinetic (area 2v, PIVC) and somatosensory stimuli (PIVC, area 3a). These three regions form the inner cortical vestibular circuit. Besides these polysensory vestibular cortical fields, three other circumscribed cortical regions of the macaque brain were also found to project directly to the brainstem vestibular nuclei: a circumscribed part of the postarcuate premotor cortex (area 6pa), part of the agranular and the adjacent dysgranular cortex located around the cingulate sulcus (area 6c/23c), and a predominantly visual (optokinetic) association field located at the fundus of the lateral sulcus (area T3). These areas are known to have connections with the structures of the inner cortical vestibular circuit. Only a few efferent connections to the brainstem vestibular nuclei were found for the different parts of cytoarchitectonic area 7. Significant differences were found between the efferent innervation patterns of the axons originating in the six cortical areas mentioned and ending in the various compartments of the vestibular nuclear complex. Vestibular nuclei with a dominant output to the gaze motor system of the brainstem receive efferent connections preferably from the parietoinsular vestibular cortex. Vestibular structures with their primary output to skeletomotor centers, however, receive stronger efferent connections from areas 6pa and 3a. The ventrolateral nucleus, which sends efferent axons to both the oculomotor and skeletomotor systems of the brainstem and the spinal cord, also receives its main cortical efferents from the somatomotor area 6 and from area 3aV. Through these connections the cortical somatomotor system may directly influence vestibuloocular and vestibulocollic reflexes. It is speculated that the corticofugal connections to the vestibular brainstem nuclei are predominantly inhibitory, suppressing vestibular reflexes during cortically controlled goal-directed movements.

Connections from the neocortex to the vestibular brain stem nuclei in the common marmoset.

Monosynaptic projections from the cerebral cortex to the vestibular nuclei were studied in the common marmoset monkey (Callithrix jacchus) by injecting fluorescent dextrans into the brain stem vestibular nuclei. The injection sites were determined by single unit vestibular responses identified later histologically. During the recordings sinewave rotation in pitch, roll or yaw or steady tilt was applied. The most notable loci of labelling were found inside the primary sensory cortex, the cortex deep down along the posterior lateral sulcus, the premotor region and the anterior cingulate cortex. From studies in other primates these cortical areas are known to process vestibular information. Their connections with the vestibular nuclei may serve as an internal feedback modulating the vestibular brain stem activity.

Corticofugal projections to the vestibular nuclei in squirrel monkeys: further evidence of multiple cortical vestibular fields.

Single- and multiple-unit recordings were made from nerve cells located in the different nuclei of the brainstem vestibular nuclear complex (VNC) of anaesthetized squirrel monkeys (Saimiri sciureus) by conventional stereotaxic techniques. After neurons responding to semicircular canal stimulation in a yaw, roll, or pitch direction or to otholith stimulation were identified, small amounts of retrograde tracer substances were deposited at the recording sites. Up to three different tracers were administered to different parts of the VNC in the same animal (Fast Blue, HRP-WGA, and Rhodamine-dextranes). After adequate survival times, the animals were sacrificed. Following histological processing, the cortical grey matter was screened systematically for cells labelled with the retrograde tracers (fluorescence microscopy or light microscopy for HRP processing). Labelled nerve cells which clearly project to the VNC directly were found predominantly in the cytoarchitectonic layer 5 of seven different cortical areas: 1) The parieto-insular vestibular cortex PIVC, which in squirrel monkeys consists mainly of the medial area Ri and parts of the anterior area Ig; 2) area 7ant, which presumably corresponds to the macaque area 2v; 3) area 3aV, a vestibular field of area 3a; 4) the temporal area T3 bordering on area Ri; 5) the premotor area 6a; and 6, 7) the areas 6c and 23c of the anterior cingulate cortex. The PIVC, area 7ant, and area 3aV form the "inner cortical vestibular circuit" (Guldin et al.: J. Comp. Neurol. 326:375-401, '92), while the other cortical areas mentioned also have direct projections to the structures of the inner cortical vestibular circuit. It is speculated that the direct projections of the cortical vestibular structures to the brainstem vestibular nuclei regulate the vestibulo-ocular, the vestibulo-spinal, and the optokinetic reflexes mediated through the VNC, thus preventing counteractions of these reflexes during voluntary, goal-directed head movements or locomotion.

Cortico-cortical connections and cytoarchitectonics of the primate vestibular cortex: a study in squirrel monkeys (Saimiri sciureus).

The cortical connections of two vestibular fields [parieto-insular vestibular cortex (PIVC) and area 3aV] were studied in the squirrel monkey (Saimiri sciureus) by means of retrograde tracer techniques. Small iontophoretic or pressure injections of horseradish peroxidase (HRP), wheat-germ-HRP, Nuclear Yellow, and Fast Blue were administered to the cytoarchitectonic areas Ri (PIVC), 3aV, the parieto-temporal association area T3, the granular insula (Ig), and the rostral part of area 7 (7ant). The injection sites were physiologically characterized by means of microelectrode recordings and vestibular, optokinetic, or somatosensory stimulation: Area Ri is the region of the parieto-insular vestibular cortex (PIVC) as defined in macaques. The neck-trunk region of area 3a (area 3aV) also contains many neurons responding to stimulation of semicircular canal receptors. Some neurons of area T3 bordering on the PIVC also receive vestibular signals, but most neurons in area T3 responded preferentially to large-field optokinetic stimulation and not to vestibular stimulation. In none of the areas mentioned were responses to otolith stimulation found. The PIVC receives inputs from frontal and parietal cortical areas, especially areas 8a, 6, 3a, 3aV, 2, and 7ant. Area T3 receives signals from the insular and retroinsular cortex, various parts of area 7, visual areas of the parieto-occipital and parieto-temporal regions (area 19) and from a sector of the upper bank of the temporal sulcus (STS-area). The cortical afferents to area 3aV stem from areas 24, 4, 6, 7ant, from other parts of the primary somatosensory cortex, the secondary somatosensory cortex (SII), the retroinsular cortex (Ri), and the granular insula (Ig). In the border region of the areas 2 and 7ant, labelled neurons appeared after injections into both the PIVC and the area 3aV. This region is presumably the homologue to the vestibular area 2v of the macaque brain. In all regions cells within the contralateral cortex were less frequently labelled than cells in the homologous structures of the ipsilateral hemisphere. The cortical system for processing vestibular information about head-in-space movement consists mainly of the reciprocally interconnected areas PIVC and 3aV, and most likely of border regions of area 2 and 7ant. This "inner cortical vestibular circuit" also receives signals from two other cortical sensory systems, the somatosensory-proprioceptive system mediated by the primary somatosensory cortex and the visual movement system (optokinetic or visual flow signals). These visual movement signals reach PIVC via area 19 and area T3.(ABSTRACT TRUNCATED AT 400 WORDS)

Thalamic connections of the vestibular cortical fields in the squirrel monkey (Saimiri sciureus).

The afferent thalamic connections to cortical fields important for control of head movement in space were analysed by intracortical retrograde tracer injections. The proprioceptive/vestibular area 3aV, the neck-trunk region of area 3a, receives two thirds of its thalamic projections from the oral and superior ventroposterior nucleus (VPO/VPS), which is considered as the proprioceptive relay of the ventroposterior complex (Kaas et al., J. Comp. Neurol. 226:211-240, 1984). The parieto-insular vestibular cortex (PIVC, area retroinsularis, Ri) receives its main thalamic input from posterior parts of the ventroposterior complex and from the medial pulvinar. Anatomical evidence is presented that the posterior region of the ventroposterior complex is a special compartment within this principal somatosensory relay complex. The parietotemporal association area T3, mainly involved in visual-optokinetic signal processing, receives a substantial input from the medial, the lateral, and the inferior pulvinar. Dual tracer experiments revealed that about 5% of the thalamic neurons projecting to 3aV were spatially intermingled with neurons projecting to areas PIVC or T3. This spatial intermingling was distributed over small but numerous, circumscribed thalamic regions, called "common patches," which were found mainly in the intralaminar nuclei, the posterior group of thalamic nuclei, and the caudal parts of the ventroposterior complex. The "common patches" may indicate a functional coupling of area 3aV with the PIVC or area T3 on the thalamic level. In control experiments thalamic projections to the granular insula Ig and the anterior part of area 7, two cerebral structures connected with the vestibular cortical areas, were studied. Some overlap in the thalamic relay structures projecting to these areas with those projecting to the vestibular cortices was found. A quantitative evaluation of thalamic regions projecting to different cortical structures was performed by constructing so-called "thalamograms." A scheme was developed that describes the afferent thalamic connections by which vestibular, visual-optokinetic, and proprioceptive signals reach the vestibular cortical areas PIVC and 3aV.

Responses of single neurons in the parietoinsular vestibular cortex of primates.

1. Neurons activated by stimulation of the horizontal and/or vertical vestibular semicircular canals were recorded in the parietoinsular vestibular cortex in four awake Java monkeys (Macaca fascicularis) and three squirrel monkeys (Saimiri sciureus). Steady tilt in darkness or during illumination of a vertically striped cylinder or of the normal laboratory surroundings did not lead to a significant change in PIVC neuron activity. Thus vestibular input to this cortical region seems to be restricted to signals originating in the semicircular canal receptors. 2. Vestibular stimulation in the three main experimental planes (roll, yaw, and pitch) and in planes in between provided clear evidence that optimum activation can be found in planes that do not coincide with the planes of the semicircular canals but are distributed over all possible spatial planes through the head. 3. Definite evidence of clustering in subdivisions of PIVC of neurons responding to the same optimum rotation plane was obtained in squirrel monkeys and is also suggested to exist in PIVC of Java monkeys. 4. Nearly all neurons responding to vestibular stimulation were also activated by visual large-field movement (optokinetic stimulation). Responses to optokinetic stimuli were always at optimum when the direction of the movement pattern corresponded to the optimum vestibular plane. Two classes of visual-vestibular interaction were found: Synergistic neurons were those PIVC cells with the strongest response to visual movement stimulation in the opposite direction to that leading to a maximum response to vestibular stimulation. Antagonistic neurons had a response maximum when the visual stimulus was moved in the direction of optimum vestibular stimulation. 5. Most PIVC neurons responded to stimulation of the deep mechanoreceptors in the neck region. This input from the neck receptors was tested quantitatively only in the horizontal plane (trunk rotation with the head fixed in space or head rotation with the trunk fixed in space). It interacted with vestibular signals at the PIVC neurons either in an antagonistic or a synergistic manner, the latter meaning activation during rotation of the head in the same direction as that leading to activation induced by semicircular canal stimulation. 6. In addition to the direction-specific vestibular, visual, and neck receptor inputs, a rather complex somatosensory input to PIVC neurons exists, including responses to stimulation of mechanoreceptors of the skin, the muscles, and the joint receptors of legs and arms. Total body vibration also led to activation of some of the neurons.(ABSTRACT TRUNCATED AT 400 WORDS)

Cortical projections originating from the cat's insular area and remarks on claustrocortical connections.

The cortical projections originating in the cat's insular cortex and claustrum were investigated with the aid of the horseradish peroxidase retrograde tracing technique. Twenty small injections of horseradish peroxidase were distributed along lateral and medial regions of the hemisphere. Labeling in the insular cortex occurred following all injections except those six situated along the lateral gyrus--that is, within the visual cortex. In the claustrum labeled neurons were found following all injections, except following the injection situated in the posterior temporal area. Claustral labeling was frequently more intense than insular labeling. The injections into the occipital cortex that revealed no insular innervation nevertheless received a considerable number of claustral projections. As the insular cortex itself receives at most a minor projection from the claustrum the differing cortical projection patterns of insula and claustrum have to be considered unrelated. Our findings confirm the view that the claustrum projects to most regions of the cerebral cortex; these projections are at least in part topographically organized. A topographical pattern can also be constructed for the insular cortex, though it is less stringent than for the claustrocortical connections. Both the afferent and efferent connections of the insula show similarities to those of the prefrontal cortex. Nevertheless, the insula differs in that it receives strong input from the sensory associative nuclei of the thalamus. Consequently, and in line with behavioral observations following its ablation, we consider the insula as involved in the temporal structuring of perceived patterns.

Cortical and thalamic afferent connections of the insular and adjacent cortex of the cat.

The thalamo-cortical and cortico-cortical afferents of the cat's insular cortex were investigated with the retrograde horseradish peroxidase technique. The most prominent loci of thalamic labeling were the suprageniculate nucleus and parts of the posterolateral nucleus. Injections into the anterior part of the insular cortex also resulted in labeled cells in the ventromedial posterior nucleus and in the intralaminar nuclei, while injections into posterior parts revealed projections from the medial and dorsal parts of the medial geniculate nucleus. Only the anterior and most ventral parts of the insular cortex overlying the anterior rhinal sulcus were connected with the mediodorsal nucleus of the thalamus. All injections into the gyrus sylvius anterior showed a specific pattern of cortical afferents: With the exception of the labeling in the prefrontal cortex and the inferotemporal region, the labeled cells were very narrowly restricted to the presylvian, the suprasylvian, and the splenial sulcus. The thalamic neurons projecting to the cortex were generally organized in a bandlike pattern which crossed nuclear borders. The majority of the cortico-cortical connections originated from sulcal areas next to the prefrontal, parietal, and cingulate cortex, that is, next to so-called association cortices. In the light of the present results the role of the insular cortex as a multifunctional association area is discussed, as well as its relation to other cortical centers.

Afferents to the ventral tegmental nucleus of Gudden in the mouse, rat, and cat.

Afferents to the ventral tegmental nucleus of Gudden (VT) were investigated in mice, rats, and cats. Unilateral and bilateral injections or iontophoretical applications of horseradish peroxidase (HRP) were made into the region of the VT. The entire cerebrum was then screened for labeled neurons. Following injections situated principally within the VT, in all three species many retrogradely labeled neurons were observed in the mamillary bodies and the lateral habenular nuclei. Fewer labeled cells were observed in the prefrontal cortex, the basal forebrain, various hypothalamic nuclei, the interpeduncular nucleus, nucleus of the posterior commissure, nucleus of Darkschewitsch and interstitial nucleus of Cajal, vestibular nucleus, and nucleus praepositus hypoglossi. Scant but consistent labeling occurred in the cingular, retrosplenial, and insular cortices, within the medial forebrain bundle, fields of Forel, zona incerta, ventral tegmental area of Tsai, substantia nigra, pretectal area, periaqueductal gray, dorsal tegmental nucleus, locus ceruleus, and raphe complex. Our results show a high similarity in the distribution of afferent connections converging on the VT of mice, rats, and cats. They indicate furthermore that the VT is reached by a variety of cortical and subcortical afferents, which belong either to the limbic system or to brain stem regions related to motor, sensory, and autonomic functions. It is suggested that the VT subserves as a midbrain core structure of the limbic system, which is responsible for the transfer of motor, sensory, and autonomic informations arising within the brain stem to limbic forebrain structures.

Heterotopic interhemispheric cortical connections in the rat.

In the rat, contrary to other species, interhemispheric cortical connections have been considered to travel largely between homotopical regions only. Based on iontophoretic injections of horseradish peroxidase, the present study reports extensive heterotopic interhemispheric connections between posterior insular (perirhinal) regions of the lateral part of the hemisphere and anterior prefrontal regions of the medial hemisphere and vice versa. Generally, the areas connected interhemispherically are also connected intrahemispherically. The ratio of contralateral projections appears to be less than one third of the ipsilateral ones.