Basal ganglia and limbic system pathology in schizophrenia. A morphometric study of brain volume and shrinkage

The volume of several parts of the basal ganglia and of the limbic system was measured by planimetry of myelin-stained serial sections in postmortem brains of 13 schizophrenic patients and nine control cases. The medial limbic structures of the temporal lobe (amygdala, hippocampal formation, and parahippocampal gyrus) and the pallidum internum were significantly smaller in the schizophrenic group, whereas the pallidum externum showed only a modest trend toward volume reduction. The volumes of the putamen, nucleus caudatus, nucleus accumbens, and the bed nucleus of the stria terminalis did not differ between patients and controls. The volume reductions of the limbic temporal structures and of the pallidum internum of schizophrenics are interpreted as degenerative shrinkages of unknown etiology.     

Changes in adult brain and behavior caused by neonatal limbic damage: implications for the etiology of schizophrenia

We tested the hypothesis that limbic damage in early development can cause aberrant maturation of brain structures known to be abnormal in adult schizophrenics: the hippocampus, prefrontal cortex, ventricles, and forebrain dopamine systems. We measured brain morphology, locomotor response to apomorphine, and cognitive processes in adult rats which received electrolytic damage to amygdala or hippocampus 48 h after birth. The behavioral measurements involved tasks which depend upon the integrity of the hippocampus or prefrontal cortex, and a task sensitive to forebrain dopamine system activation. The tasks included place navigation, egocentric spatial ability, and apomorphine-induced locomotion. The rats with lesions showed poor performance on the place navigation and egocentric spatial tasks and more apomorphine-induced locomotion after puberty than the sham lesion group. Regardless of lesion location, the adult rats showed smaller amygdalae and hippocampi, and larger lateral ventricles. Analyzing the lesion and sham rats together, adult amygdala volume was found to be positively correlated with cerebral cortex, prefrontal cortex, and hippocampal volumes and place navigation performance, and was negatively correlated with lateral ventricle volume. This study contributes to our understanding of the pathogenesis of schizophrenia by showing that early damage to limbic structures produced behavioral, morphological, and neuropharmacological abnormalities related to pathology in adult schizophrenics.     

Reduced anterior hippocampal formation volume in hyponatremic schizophrenic patients

Diminished hippocampal volume occurs in the anterior segment of some schizophrenic patients, and in the posterior segment in others. The significance of hippocampal pathology in general and these segmental differences in specific is not known. Several lines of evidence suggest anterior hippocampal pathology underlies the life-threatening hyponatremia seen in a subgroup of patients with schizophrenia; therefore our goal was to determine if this region was preferentially diminished in hyponatremic patients. We studied seven polydipsic hyponatremic, ten polydipsic normonatremic, and nine nonpolydipsic normonatremic schizophrenic inpatients, as well as 12 healthy controls. All underwent structural scanning on a high resolution (3.0 T) magnetic resonance imaging (MRI) scanner. Hippocampal formation, amygdala, and third ventricle volumes were manually traced in each subject. The hippocampus was divided at the posterior extent of the uncus, and all structural volumes were corrected for whole brain volume and other significant recognized factors (i.e., age, gender, height, parental education). Despite being overhydrated, anterior hippocampal formation volume was diminished in those with polydipsia and hyponatremia relative to each of the other three groups. Third ventricle volume was larger in this group than in healthy controls but similar to the two patient groups. Posterior hippocampal and amygdala volumes did not differ between groups. Other potential confounds (e.g., water imbalance) either had no effect or accentuated these differences. We conclude the anterior hippocampal formation is smaller in hyponatremic schizophrenic patients, thereby linking an important and objective clinical feature of schizophrenia to a neural pathway that can be investigated in animal models. The findings strengthen the hypothesis that anterior hippocampal formation pathology disrupts functional connectivity with other limbic structures in schizophrenia.     

Structural pathology underlying neuroendocrine dysfunction in schizophrenia

Polydipsic hyponatremic schizophrenic (PHS) patients exhibit altered neuroendocrine activity that has been linked to their life-threatening water imbalance, as well as to impaired function and reduced volume of the anterior hippocampus. Polydipsic patients without hyponatremia (polydipsic normonatremic schizophrenics: PNS) exhibit similar, albeit less marked, changes in neuroendocrine activity and anterior hippocampal function, but not reduced anterior hippocampal volume. Indeed, reduced anterior hippocampal volume is seen in patients with normal water balance (nonpolydipsic normonatremic schizophrenics: NNS) whose neuroendocrine activity and anterior hippocampal function differ markedly from those with polydipsia. In an effort to reconcile these findings we measured hippocampal, amygdala and 3rd ventricle shapes in 26 schizophrenic patients (10 PNS, 7 PHS, 9 NNS) and 12 healthy controls matched for age and gender. Bilateral inward deformations were localized to the anterior lateral hippocampal surface (part of a neurocircuit which modulates neuroendocrine responses to psychological stimuli) in PHS and to a lesser extent in PNS, while deformations in NNS were restricted to the medial surface. Proportional deformations of the right medial amygdala, a key segment of this neurocircuit, were seen in both polydipsic groups, and correlated with the volume of the 3rd ventricle, which lies adjacent to the neuroendocrine nuclei. Finally, these structural findings were most marked in those with impaired hippocampal-mediated stress responses. These results reconcile previously conflicting data, and support the view that anterior lateral hippocampal pathology disrupts neuroendocrine function in polydipsic patients with and without hyponatremia. The relationship of these findings to the underlying mental illness remains to be established.     

Neonatal lesions of the left entorhinal cortex affect dopamine metabolism in the rat brain

The present study was performed to determine the effects of neonatal excitotoxic lesions of the left entorhinal cortex on dopamine (DA) metabolism and release in limbic regions of the rat brain. Quinolinic acid or phosphate buffered saline was infused into the left entorhinal cortex of rat pups on postnatal day 7 (PD7). Concentrations of DA,3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) in the lateral amygdala, nucleus accumbens, caudate-putamen, and medial prefrontal cortex were determined in the postmortem brains of lesioned and sham-operated rats on PD35 and PD56. On PD35, concentrations of DA in the bilateral lateral amygdala and HVA in the left lateral amygdala were significantly increased in lesioned rats compared with sham-operated animals, while no significant change was observed in the other three brain areas. On PD56, in addition to the increased concentration of DA in the left lateral amygdala, those of DA, DOPAC and HVA in the caudate-putamen, and DA in the nucleus accumbens were found to be increased, but DA concentrations in the right medial prefrontal cortex were decreased. The DOPAC/DA concentration ratio was, however, decreased in the amygdala and nucleus accumbens of the lesioned rats. In an in vivo microdialysis study, methamphetamine (MAP: 2 mg/kg, i.p.)-induced DA release in the amygdala of lesioned rats was significantly enhanced compared with sham-operated rats on both PD35 and PD56. There were no significant differences in MAP-induced DA release in the caudate-putamen between the sham-operated and lesioned rats at any time point. These findings provide evidence that neonatally induced structural abnormalities in the entorhinal cortex affect DA transmission in the limbic regions at the adolescent stage.     

Topographically specific hippocampal projections target functionally distinct prefrontal areas in the rhesus monkey

The sources of ipsilateral projections from the hippocampal formation, the presubiculum, area 29a-c, and parasubiculum to medial, orbital, and lateral prefrontal cortices were studied with retrograde tracers in 27 rhesus monkeys. Labeled neurons within the hippocampal formation (CA1, CA1′, prosubiculum, and subiculum) were found rostrally, although some were noted throughout the entire rostrocaudal extent of the hippocampal formation. Most labeled neurons in the hippocampal formation projected to medial prefrontal cortices, followed by orbital areas. In addition, there were differences in the topography of afferent neurons projecting to medial when compared with orbital cortices. Labeled neurons innervating medial cortices were found mainly in the CA1′ and CA1 fields rostrally, but originated in the subicular fields caudally. In contrast, labeled neurons which innervated orbital cortices were considerably more focal, emanating from the same relative position within a field throughout the rostrocaudal extent of the hippocampal formation. In marked contrast to the pattern of projection to medial and orbital prefrontal cortices, lateral prefrontal areas received projections from only a few labeled neurons found mostly in the subicular fields. Lateral prefrontal cortices, lateral prefrontal cortices received the most robust projections from the presubiculum and the supracallosal area 29a-c. Orbital, and to a lesser extent medial, prefrontal areas received projections from a smaller but significant number of neurons from the presubiculum and area 29a-c. Only a few labeled neurons were found in the parasubiculum, and most projected to medial prefrontal areas. The results suggest that functionally distinct prefrontal cortices receive projections from different components of the hippocampal region. Medial and orbital prefrontal cortices may have a role in long-term mnemonic process similar to those associated with the hippocampal formation with which they are linked. Moreover, the preponderance of projection neurons from the hippocampal formation innervating medial when compared with orbital prefrontal areas followed the opposite trend from what we had observed previously for the amygdala (Barbas and De Olmos [1990]) (J Comp Neurol 301:1–23). Thus, the hippocampal formation, associated with mnemonic processes, targets predominantly medial prefrontal cortices, whereas the amygdala, associated with emotional aspects of memory, issues robust projections to orbital limbic cortices. Lateral prefrontal cortices receive robust projections from the presubiculum and area 29a-c and sparse projections from the hippocampal formation. These findings are consistent with the idea that the role of lateral prefrontal cortices in memory is distinct from that of either medial or orbital cortices. The results suggest that signals from functionally distinct limbic structures to some extent follow parallel pathways to functionally distinct prefrontal cortices. © 1995 Wiley-Liss, Inc.     

Increased frequency of dentate granule cells with basal dendrites in the hippocampal formation of schizophrenics

In the hippocampal formation of schizophrenics, the detailed morphology of Golgi-impregnated granule cells was examined. These granule cells of the dentate gyrus are interposed between the rostral entorhinal cortex and the hippocampus proper. In these limbic regions significant cytoarchitectural alterations in schizophrenics are reported, giving rise to the concept of a prenatal limbic maldevelopment in schizophrenia. Compared to controls, the frequency of dentate granule cells with basal dendrites was significantly increased in schizophrenics [43% (+/-3)] vs. [22% (+/-2) in the control group]. In epilepsy, dentate granule cells of epileptic patients also develop basal dendrites, which is explained as an adaptive process of plasticity. Similarly, the hippocampal alterations described in schizophrenics could be the sequela of primary entorhinal cytoarchitectural alterations. Since the increase in basal dendrites seems to reflect a process of continuous plasticity, suggesting an increased rate of postnatal granule cell generation, the synthesis of a prenatal limbic maldevelopment with an ongoing process of plasticity might, therefore, supersede the hypothesis of a neurodegeneration in schizophrenia.     

Cortex, white matter, and basal ganglia in schizophrenia: a volumetric postmortem study

Postmortem volumetry of cortex, white matter, and basal ganglia was performed in 23 brains of schizophrenic patients and 23 brains of controls closely matched for gender, age, and hemisphere. Stereological methods were applied to serial coronal sections of complete hemispheres. We found no significant volume changes of cortex and white matter in schizophrenics. Striatal volume of schizophrenics was increased bilaterally reaching a significant level on the left side. Volumes of the globus pallidus were increased in both hemispheres reaching a significant level on the right side. After psychopathological differentiation, basal ganglia volume increase was also found in the subgroup of paranoid-hallucinatory schizophrenics.     

Anterior hippocampal volume reduction in male patients with schizophrenia

Quantitative high resolution magnetic resonance imaging (MRI) was utilized to measure anterior, posterior, and total hippocampal volumes in 27 male patients with chronic schizophrenia and 24 male controls. To optimize measurement techniques, hippocampal volumes were: (1) acquired with 1.4-mm slices; (2) excluded with the amygdala; (3) normalized for position; and (4) corrected for total intracranial volume (ICV). The results of a linear mixed effects regression analysis, which made it possible to analyze total anterior and total posterior hippocampal volumes separately, indicated that the anterior hippocampus was significantly smaller in the schizophrenic group relative to the control group. There were no significant group differences with respect to posterior hippocampal volumes, and no significant correlations between hippocampal volumes and illness duration. A significant lateralized asymmetry was also noted in both groups with the right hippocampal volume being larger than the left. These preliminary findings support a significant anterior hippocampal volume reduction in men with schizophrenia as well as a similar hippocampal volume asymmetry in both male controls and schizophrenics.     

Effects of hippocampal and amygdalar stimulation on uptake and binding of 3H-hydrocortisone in the hypothalamus of the cat

Stimulation of the hippocampus or amygdala of adrenalectomized cats occurred for 10 sec followed by a 50 sec period of no stimulation, beginning 30 min prior to and ending 30 min after administration of 100 μCi of ³H-hydrocortisone into a lateral ventricle. Sixty min after administration of labeled hormone, the hypothalamus was excised and homogenized. Cytosol and nuclear extract fractions were obtained and analyzed for radioactivity and protein content. Separation of bound from free hydrocortisone was achieved by charcoal adsorption assay. Results reveal that stimulation of the hippocampus resulted in a greater concentration of ³H-hydrocortisone taken up into hypothalamic cells. Also, a greater percentage of total hormone found in the nuclear extract was assayed as bound ³H-hydrocortisone, and the concentration of bound radioactivity in the nuclear extract was increased over control values. Amygdalar stimulation, in general, yielded results similar to those obtained from control cats. However, although a lesser percentage of total hormone in the hypothalamic cytosol was assayed as bound hormone, there was a greater concentration of nuclear bound hormone than in controls, but less than that determined in the hippocampal stimulation group. These results add to the evidence that hippocampus and amygdala have a modulating influence upon the hypothalamo-hypophyseal axis. They also suggest that one manner in which these limbic structures may influence hypothalamic function is to modulate the uptake and binding of hydrocortisone in hypothalamic cells.     

Cell loss in the hippocampus of schizophrenics

Summary To investigate whether volume reduction of the hippocampal formation of schizophrenics, as described previously, is paralleled by loss of neurons and fibre systems, tissue volumes and cell numbers of all parts of the hippocampal formation in post mortem brains of 13 schizophrencis and 11 agematched controls belonging to the Vogt collection were determined.Volumes of the whole hippocampal formation (P < 0.01), the whole pyramidal band (P < 0.001) and the hippocampal segments CA1/CA2 (P < 0.01), CA3 (P < 0.05), CA4 (P < 0.01) were decreased, whereas no significant volume reduction of the alveus and fimbria hippocampi and presubiculum/subiculum could be found. The perforant path showed a trend towards volume reduction (P < 0.1).The absolute number of pyramidal cells (tissue volume × cell density) was diminished in CA1/CA2 (P < 0.05), CA3 (P < 0.05) and CA4 (P < 0.05), but was not significantly changed in the prosubiculum/subiculum, the presubiculum/parasubiculum and the granular cell layer of the dentate fascia.Pyramidal cell loss in CA1/CA2, CA3, CA4 was more distinct in the paranoid patients than in catatonics. The findings are discussed with respect to current hypotheses of limbic dysfunction in schizophrenia.Rheinische Landesklinik, Psychiatrische Klinik der Universität Düsseldorf, Bergische Landstrasse 2, D-4000 Düsseldorf 12 Federal Republic of Germany     

Frontolimbic brain abnormalities in patients with borderline personality disorder: a volumetric magnetic resonance imaging study.

BACKGROUND: Dual frontolimbic brain pathology has been suggested as a possible correlate of impulsivity and aggressive behavior. One previous study reported volume loss of the hippocampus and the amygdala in patients with borderline personality disorder. We measured limbic and prefrontal brain volumes to test the hypothesis that frontolimbic brain pathology might be associated with borderline personality disorder. METHODS: Eight unmedicated female patients with borderline personality disorder and eight matched healthy controls were studied. The volumes of the hippocampus, amygdala, and orbitofrontal, dorsolateral prefrontal, and anterior cingulate cortex were measured in the patients using magnetic resonance imaging volumetry and compared to those obtained in the controls. RESULTS: We found a significant reduction of hippocampal and amygdala volumes in borderline personality disorder. There was a significant 24% reduction of the left orbitofrontal and a 26% reduction of the right anterior cingulate cortex in borderline personality disorder. Only left orbitofrontal volumes correlated significantly with amygdala volumes. CONCLUSIONS: While volume loss of a single brain structure like the hippocampus is quite an unspecific finding in neuropsychiatry, the patterns of volume loss of the amygdala, hippocampus, and left orbitofrontal and right anterior cingulate cortex might differentiate borderline personality disorder from other neuropsychiatric conditions.     

Magnetic resonance imaging findings in schizophrenia and atypical psychoses

The differences among MRI findings were studied in schizophrenic psychoses. The schizophrenics and atypical psychotics had significant reductions in bilateral hippocampal volumes compared to controls, but the two patient groups did not differ from each other. As for ventricle volume, the schizophrenics showed significantly larger temporal horns and third ventricle than normal controls, whereas atypical psychotics did not. Moreover, the left temporal horn in the schizophrenics was significantly larger than that seen in the atypical psychotics. By cluster analysis, schizophrenics and atypical psychotics were found to have a tendency to be distributed in different groups. These results might be considered to support the classification of schizophrenic psychoses into schizophrenia and atypical psychoses.     

Limbic influence on the periaqueductal gray: A single unit study in the awake squirrel monkey

Single neurons of the periaqueductal gray (PAG) were studied during electrical stimulation of the amygdala and hippocampus. Fifty-one percent (34/67) of the units sampled throughout the rostrocaudal extent of the PAG were found to have a limbic influence. PAG neurons were characterized by low spontaneous firing rates (X¯= 4.94 spikes/sec). Units responded to basolateral amygdala stimulation primarily with short duration excitatory responses having a mean latency of 30 ms (range: 13.3–110 ms). Responses to corticomedial and lateral amygdala stimulation produced different patterns of activation including complex excitatory and inhibitory sequences. Only 10 units (15%) sampled in PAG responded to hippocampal stimulation with excitatory or tonic-inhibitory responses. The majority of responsive units (8) were to anterior hippocampal stimulation (latency range: = 20–75 ms). High frequency (9 Hz) basolateral amygdala stimulation recruited responses with increases in the probability of firing and a decrease in initial latency and latency variability.     

The effect of limbic and extralimbic electrical stimulations upon prolactin secretion in humans

The effect that extra-hypothalamic regions of the brain have upon prolactin secretion in humans was evaluated by performing electrical stimulations. Thirty-nine stimulations were performed, 22 to basolateral amygdala, 12 to hippocampus and 5 to orbitofrontal, supplementary motor and cingulate cortex. Only two stimulations causing high-frequency widespread limbic afterdischarges were followed by significant prolactin elevation. Four low-frequency afterdischarges involving amygdala and anterior hippocampus, one amygdala stimulus-dependent discharge and 19 amygdala, 8 hippocampal and 5 frontal sub-afterdischarge threshold stimulations had no prolactin elevation. These results fail to replicate earlier studies. We suggest that there is no evidence that the amygdala regulates serum prolactin within physiologic ranges, but that the regulation of prolactin may depend primarily upon other sub-cortical structures.     

Amygdalar and hippocampal MRI volumetric reductions in Parkinson's disease with dementia.

Parkinson's disease (PD) involves neuropathological changes in the limbic system that lead to neuronal loss and volumetric reductions of several nuclei. We investigated possible volumetric reductions of the amygdala and hippocampus associated to PD. We carried out magnetic resonance imaging (MRI) volumetric studies in 16 patients with PD and dementia (PDD), 16 patients with PD without dementia (PD), and 16 healthy subjects. The general analysis of variance (ANOVA) showed a significant group effect (for the amygdala, P = 0.01; for the hippocampus, P = 0.005). A post-hoc test demonstrated that the differences were due to PDD and control group comparisons for the amygdala (P = 0.008) and for the hippocampus (P = 0.004). In nondemented PD subjects, we observed an 11% reduction in the amygdala and a 10% reduction in the hippocampus compared with that in controls. In summary, demented PD patients have clear amygdalar and hippocampal atrophy that remains statistically significant after controlling for global cerebral atrophy. Nondemented PD patients also showed a degree of volumetric reduction in these structures although the differences were not statistically significant.     

Post-mortem volume measurements of limbic system and basal ganglia structures in chronic schizophrenics. Initial results from a new brain collection

Volumes of the hippocampal formation, external and internal pallidum, caudate, putamen, and nucleus accumbens were measured in both hemispheres of recently collected post-mortem brains of 18 chronically ill schizophrenics and 21 control subjects. In the schizophrenic group, the hippocampal formation and the internal pallidum were significantly smaller in the right and left hemisphere, whereas external pallidum, putamen, caudate and accumbens were not significantly changed. Volumes of the hippocampus and of all evaluated parts of the basal ganglia were in the male schizophrenics more reduced than in the female patients. The right and left hemispheres were equally affected in both sexes. Since the mean brain weight was in patients and controls nearly identical, the volume differences can not be explained by a general brain atrophy or hypoplasia but rather indicate a more focal lack of brain tissue, by which some clinical features of the disease might be explained.     

Changes in gray-/white-matter ratios in the parahippocampal gyri of late-onset schizophrenia patients.

OBJECTIVE: Recent neuropathological studies have reported that late-onset schizophrenia patients exhibit a restricted limbic tauopathy, glial tangles (thorn-shaped astrocytes), scarce amyloid deposition, and preservation of hippocampal pyramidal cells. The present article is an attempt at finding a macroscopic correlate to the described pathology. METHODS: A group of 13 normal-onset (<40 years) schizophrenic patients, 13 late-onset (>40 years) schizophrenia patients, and 8 comparison clients were studied, based on fulfillment of diagnostic criteria (e.g., DSM-III) and availability of suitable tissue. A computerized image analysis provided areal measurements of the hippocampal formation. Both hemispheres and two levels (mammillary bodies and lateral geniculate nucleus) were studied. In order to avoid corrections based on tissue shrinkage, results were expressed as ratios (e.g., parahippocampal gray/white matter). RESULTS: Late-onset schizophrenic patients exhibit significant alterations in the gray-/white-matter ratio of the parahippocampal gyrus, affecting both hemispheres and all levels examined. This finding can best be explained by the preservation of gray matter and concomitant reduction of white matter in affected parahippocampal gyri. CONCLUSION: The presence of neuritic changes, preservation of pyramidal cell numbers, and of areal diminution of the parahippocampal white matter can best be explained by a dying-back neuropathy.     

MRI volumes of amygdala and hippocampus in non-mentally retarded autistic adolescents and adults

OBJECTIVE: To determine whether volumes of hippocampus and amygdala are abnormal in people with autism. BACKGROUND: Neuropathologic studies of the limbic system in autism have found decreased neuronal size, increased neuronal packing density, and decreased complexity of dendritic arbors in hippocampus, amygdala, and other limbic structures. These findings are suggestive of a developmental curtailment in the maturation of the neurons and neuropil. METHODS: Measurement of hippocampus, amygdala, and total brain volumes from 1.5-mm coronal, spoiled gradient-recalled echo MRI scans in 14 non-mentally retarded autistic male adolescents and young adults and 14 individually matched, healthy community volunteers. RESULTS: Amygdala volume was significantly smaller in the autistic subjects, both with (p = 0.006) and without (p = 0.01) correcting for total brain volume. Total brain volume and absolute hippocampal volume did not differ significantly between groups, but hippocampal volume, when corrected for total brain volume, was significantly reduced (p = 0.04) in the autistic subjects. CONCLUSIONS: There is a reduction in the volume of amygdala and hippocampus in people with autism, particularly in relation to total brain volume. The histopathology of autism suggests that these volume reductions are related to a reduction in dendritic tree and neuropil development, and likely reflect the underdevelopment of the neural connections of limbic structures with other parts of the brain, particularly cerebral cortex.     

Connections of the parahippocampal cortex in the cat. IV. Subcortical efferents

The present report deals with the projections from the entorhinal and perirhinal cortices to subcortical forebrain structures and the brainstem in the cat. By using anterograde and retrograde tracing techniques, it could be demonstrated that the entire mediolateral extent of the parahippocampal cortex issues prominent projections to the dorsal and ventral striatum, the amygdala, and the claustrum. In addition, the entorhinal cortex sends projections to the septum and the diagonal band of Broca. Only the perirhinal cortex gives rise to a weak projection to the dorsolateral periaquaductal gray and the ventral pontine region. The major proportion of the subcortical projections originates in the perirhinal cortex and the lateral entorhinal cortex, whereas the medial entorhinal cortex has a much sparser output and sends no fibers to the amygdala. The subcortical projections from both the entorhinal cortex and the perirhinal cortex arise mostly from their deep layers. It was further found that these projections are topographically organized along the mediolateral axis of the parahippocampal cortex. This mediolateral axis is related to a ventrolateral to dorsomedial axis in the septum, a mediolateral axis in the amygdala and the ventral striatum, and a ventrodorsal coordinate in the dorsal striatum and the claustrum. A further topography was observed in the projections from the perirhinal cortex to the lateral amygdaloid nucleus. A rostrocaudal axis in the perirhinal cortex corresponds to a mediolateral axis in the lateral amygdaloid nucleus. The present observations are compared with data concerning the connectivity of the parahippocampal cortex with the hippocampal formation and other cortical structures. It is suggested that the parahippocampal cortex in the cat may be conceptualized as an interface between the hippocampal formation and several subcortical structures in the realm of the limbic and motor systems.