Search

Several genes essential for neocortical layering have been identified in recent years, but their precise roles in this process remain to be elucidated. Mice deficient in p35--an activator of cyclin-dependent kinase 5 (Cdk5)--are characterized by a neocortex that has inverted layering. To decipher the physiological mechanisms that underlie this defect, we compared time-lapse recordings between p35(-/-) and wild-type cortical slices. In the p35(-/-) neocortex, the classic modes of radial migration--somal translocation and locomotion--were largely replaced by a distinct mode of migration: branched migration. Branched migration is cell-autonomous, associated with impaired neuronal-glial interaction and rare in neurons of scrambler mice, which are deficient in Dab1. Hence, our findings suggest that inside-out layering requires distinct functions of Reelin and p35/Cdk5 signaling, with the latter being important for proper glia-guided migration.

Layer formation in the developing cerebral cortex requires the movement of neurons from their site of origin to their final laminar position. We demonstrate, using time-lapse imaging of acute cortical slices, that two distinct forms of cell movement, locomotion and somal translocation, are responsible for the radial migration of cortical neurons. These modes are distinguished by their dynamic properties and morphological features. Locomotion and translocation are not cell-type specific; although at early ages some cells may move by translocation only, locomoting cells also translocate once their leading process reaches the marginal zone. The existence of two modes of radial migration may account for the differential effects of certain genetic mutations on cortical development.

Analysis of genetic mutations that lead to abnormal migration and layer formation in the developing cerebral cortex of mice and humans has led to important new discoveries regarding the molecular mechanisms that underlie these processes. Genetic manipulation and experimental analysis have demonstrated significant tangential migrations of cortical neurons, some arriving from very distant noncortical sites.

Background: Magnetic resonance imaging of the brain has been of limited usefulness in the diagnosis of Creutzfeldt-Jakob disease. Abnormalities on T2-weighted images have been described, but these are neither highly sensitive nor specific., Objective: To determine whether diffusion-weighted magnetic resonance images might be useful in the evaluation of Creutzfeldt-Jakob disease., Case Presentation: A 61-year-old woman with rapidly progressive dementia was referred for cranial magnetic resonance imaging. Diffusion-weighted images were obtained as part of the examination. Brain biopsy confirmed the diagnosis of Creutzfeldt-Jakob disease histologically., Findings and Conclusions: The diffusion-weighted magnetic resonance brain images demonstrated bilaterally symmetrical marked increase in signal intensity in the caudate nuclei, putamina, thalami, cingulate gyri, and right inferior frontal cortex. The apparent diffusion coefficient map showed abnormally low diffusion in these regions (as low as 40% of normal in the caudate head). This suggests that there is restricted diffusion in these regions. The T2-weighted images demonstrated slightly increased signal bilaterally in the caudate nuclei and putamina. These findings indicate that diffusion magnetic resonance imaging might be a sensitive means of imaging the abnormalities seen in Creutzfeldt-Jakob disease.

Proteoglycans influence axonal outgrowth in several experimental paradigms, and their distribution during development suggests a role in axon guidance. We have used a monoclonal antibody, 5D4, that recognizes an epitope on sulfated keratans (KS), to define the distribution of keratan sulfate proteoglycans (KSPGs) in the developing thalamus and cortex of the rat. During development, 5D4 immunolabeling is present on thalamic axons as they grow through the internal capsule and subplate but is not present in the adjacent pathway for cortical efferent axons. Individual thalamic nuclei differ markedly in their expression of KSPGs; these distinctions persist throughout the period of developmentally regulated expression. Major cortical domains also differ in their expression of KSPGs, which are expressed throughout medial (cingulate and retrosplenial) cortex well before neocortex. Immunolabeling for KSPGs diminishes 2 weeks after birth; in the adult it is associated with small glia. The 5D4 epitope is present on several KSPGs (320, 220, and 160 kD) on Western blots during development but only in a broad 200-kD band in adult brain. Immunolabeling is degraded on sections and Western blots by keratanase II but not by keratanase I or chondroitinase ABC, confirming that the antibody recognizes KS. Bands identified by 5D4 on Western blots differ from those identified by antibodies to known KSPGs (aggrecan, claustrin, SV2, ABAKAN, phosphacan-KS), indicating that 5D4 is labeling KSPGs not previously described in the brain. The selective expression of KSPGs during development suggests that they may be a part of the molecular identity of thalamic nuclei and cortical domains that defines their connectivity.

The marginal zone (MZ) of embryonic neocortex is crucial to its normal development. We report that neurotrophin-4 (but not NT3 or NGF), applied to embryonic rodent cortex in vitro or in vivo, produces heterotopic accumulations of neurons in the MZ. Although heterotopia production is TrkB mediated, BDNF is >10-fold less effective than NT4. Heterotopic neurons have the same birth date and phenotype as normal MZ neurons; they are not the result of NT4-induced proliferation or rescue from apoptosis. We suggest that NT4 causes excess neurons to migrate into the MZ and thus may play a role in normal MZ formation as well as in the pathogenesis of certain human cortical dysplasias.

The formation of the distinct layers of the cerebral cortex begins when cortical plate neurons take up positions within the extracellular matrix (ECM)-rich preplate, dividing it into the marginal zone above and the subplate below. We have analyzed this process in the reeler mutant mouse, in which cortical lamination is severely disrupted. The recent observation that the product of the reeler gene is an ECM-like protein that is expressed by cells of the marginal zone indicates a critical role for ECM in cortical lamination. We have found that preplate cells in normal cortex that are tagged during their terminal division with bromodeoxyuridine (BrdU) are closely associated with chondroitin sulfate proteoglycans (CSPGs), which were identified by immunolabeling; this association is maintained in the marginal zone and subplate after the preplate is divided by cortical plate formation. Cortical plate cells do not aggregate within the preplate in reeler; instead, preplate cells remain as an undivided superficial layer containing abundant CSPGs, and cortical plate neurons accumulate below them. These findings indicate that preplate cells are responsible for the formation of a localized ECM, because the association of CSPGs with preplate cells is maintained even when these cells are in abnormal positions. The failure of cortical plate neurons to aggregate within the framework of the preplate and its associated ECM and to divide it is one of the earliest structural abnormalities detectable in reeler cortex, suggesting that this step is important for the subsequent formation of cortical layers.

The cerebral cortex is distinguished by layers of neurons of different morphologies and densities. The layers are formed by the migration of newly generated neurons from the ventricular zone to the cortical plate near the outer (pial) boundary of the cortex, along radial paths approximately perpendicular to the cortical surface. Immunochemical labelling makes these cells' patterns visible in brightfield microscopy so that layer formation can be studied. We developed a suite of programs that automatically digitize the entire cortex, identify the labelled cells and compute cell densities along local radial paths. Cell identification used supervised classification on all the significantly stained objects corresponding to maxima in lowpass filtered versions of the digital microgrphs. Classification of all the stained objects as cells or noncell objects was made by a decision rule based on morphometric and grey-level texture features, including features based on Gabor functions. Detection sensitivity and classification accuracy were jointly maximized on training data consisting of about 3000 expert-identified neurons in micrographs. Total program performance was tested on a separate (test) set of labelled neurons the same size as the training data set. The program detected 85% of the cells in the test set with a total error of 0.19. The identified cells' locations were used to compute population densities along normals to the cortical layers, and these densities served as a measure of neuronal migration. Transcortical density profiles obtained by computation and by manual cell counting were very similar. The cell identification program was built on well-established methods in statistical pattern recognition and image analysis and should generalize readily to other histological preparations.

Studies of the distribution and production of ECM components during development of the cerebral cortex have suggested several hypotheses regarding their functional role. In the earliest stages of cortical development, fibronectin is produced by cells in the ventricular zone throughout the telencephalic vesicle, where it may serve as a part of the local environment that supports cell division and determines cell fate. Fibronectin is also distributed along radial glial processes. It is closely associated with preplate neurons, as are chondroitin sulfate proteoglycans and several other ECM components. This association continues as preplate cells are divided into the marginal zone and subplate by the invasion of cortical plate neurons, suggesting that ECM, preplate cells and radial glia serve as a scaffold for cortical plate formation. Fibronectin is also produced by migrating neurons, but only by those moving into specific cortical domains, suggesting that it may help neurons destined for specific targets discriminate between adjacent glial guides. A recently defined ECM-like protein, reelin, is absent or abnormal in the reeler mutant mouse in which cortical neurons are severely malpositioned. Reelin is produced by marginal zone cells and is therefore appropriately located to serve as a stop signal for migrating neurons. Axons leaving the cortical plate cross the CSPG-rich subplate, then turn to follow a path containing much less CSPG. In contrast, the cortical trajectory of thalamic axons is centered on the subplate, indicating that CSPGs in the subplate are not a barrier to axon outgrowth and may instead be serving as guidance cues that distinguish afferent from efferent pathways. Neurocan, a CNS-specific CSPG with many molecular features that indicate roles in cell-cell and cell-substrate interactions, is the only CSPG defined to date whose distribution supports a role in distinguishing afferent from efferent pathways.

The distribution of fibronectin (FN) changes rapidly during early development of the cerebral cortex, but its cellular source is not known. With in situ hybridization we find two spatially and temporally distinct periods of FN mRNA expression in the embryonic and early postnatal cortex of the mouse. Before and during formation of the preplate by the first postmitotic neurons, FN mRNA levels are high throughout the telencephalic vesicle, deep in the neuroepithelial proliferative zone that contains dividing cells and the cell bodies of radial glia; expression in the cortical proliferative zone is limited to the period of neurogenesis. Just after the cortical plate is formed within the preplate, FN mRNA is expressed in the intermediate zone, which contains migrating neurons, and in the cortical plate, where neurons migrate past their predecessors to form layers. Brefeldin A treatment of an organotypic slice preparation demonstrates FN production in the intermediate zone and cortical plate, in locations that correspond exactly to the distribution of FN mRNA by in situ hybridization. FN-producing cells immunolabel with neuron-specific markers; in the intermediate zone and lower cortical plate they have morphological features characteristic of migrating neurons and are closely apposed to radial glia. FN mRNA expression and protein production continue in neurons of the cortical plate through the period of layer formation and then are downregulated. Examination of dissociated cortical cells by laser confocal microscopy confirms that FN accumulation after brefeldin A treatment is intracellular in neurons as well as in glia. Neuroepithelial expression of FN mRNA takes place throughout the telencephalon; FN produced by neurons is restricted to cells migrating toward and into specific cortical domains that include neocortex, insular and perirhinal cortex, and subiculum. Thus FN may be involved initially in supporting the cell division and fate determination that takes place in the neuroepithelium; later production by migrating neurons may play a role in the selection of radial glial pathways that lead to specific cortical regions, and in interactions between neurons as they form cortical layers within these regions.

The first thalamocortical axons to arrive in the developing cerebral cortex traverse a pathway that is separate from the adjacent intracortical pathway for early efferents, suggesting that different molecular signals guide their growth. We previously demonstrated that the intracortical pathway for thalamic axons is centered on the subplate (Bicknese et al. [1994] J. Neurosci. 14:3500-3510), which is rich in chondroitin sulfate proteoglycans (CSPGs; Sheppard et al. [1991] J. Neurosci. 11:3928-3942), whereas efferent axons cross the subplate to exit in a zone containing much less CSPG. To define the molecular composition of the subplate further, we used antibodies against CSPG core proteins and chondroitin sulfate disaccharides in an immunohistochemical analysis of their distribution in the developing neocortex of the rat. Immunolabeling for neurocan, a central nervous system-specific CSPG (Rauch et al. [1992] J. Biol. Chem. 267:19537-19547), and for chondroitin 6-sulfate and unsulfated chondroitin becomes prominent in the subplate before the arrival of thalamic afferents. Immunolabeling is initially sparse in the cortical plate but appears later in maturing cortical layers. A postnatal decline in immunolabeling occurs uniformly for most proteoglycans, but, in the somatosensory cortex, labeling for neurocan, phosphacan, and chondroitin 4- and 6-sulfate declines in the centers of the whisker barrels before the walls. In contrast to neurocan, immunolabeling for other proteoglycans is either uniformly distributed (syndecan-1, N-syndecan, 5F3, phosphacan, chondroitin 4-sulfate), restricted to axons (PGM1), distributed exclusively on nonneuronal elements (2D6, NG2, and CD44), or undetectable (9.2.27, aggrecan, decorin). Thus, neurocan is a candidate molecule for delineating the intracortical pathway of thalamocortical axons and distinguishing it from that of cortical efferents.

To identify cell-surface molecules that mediate interactions between neurons and their environment during neural development, we used monoclonal antibody techniques to define a developmentally regulated antigen in the central nervous system of the mouse. The antibody we produced (2A1) immunolabels cells throughout the central nervous system; we analyzed its distribution in the developing cerebral cortex, where it is expressed on cells very soon after they complete mitosis and leave the periventricular proliferative zone. Expression continues into adult life. The antibody also labels the epithelium of the choroid plexus and the renal proximal tubules, but does not label neurons of the peripheral nervous system in the dorsal root ganglia. In dissociated cell culture of embryonic cerebral cortex, 2A1 labels the surface of neurons but not glia. Immunolabeling of neurons in tissue culture is particularly prominent on the edge of growth cones, including filopodia and the leading edge of lamellipodia, when observed with either immunofluorescence or freeze-etch immunoelectron microscopy. Immunopurification with 2A1 of a CHAPS-extracted membrane preparation from brains of neonatal mice produces a broad (32-36 kD) electrophoretic band and a less prominent 70 kD band that are sensitive to N-glycosidase but not endoglycosidase H. Thus the 2A1 antibody recognizes a developmentally regulated, neuronal cell surface glycoprotein (or glycoproteins) with complex N-linked oligosaccharide side chains. We have termed the glycoprotein antigen EMA because of its prominence on the edge membrane of growth cones. EMA is similar to the M6 antigen (Lagenaur et al: J. Neurobiol. 23:71-88, 1992) in apparent molecular weight, distribution in tissue sections, and immunoreactivity on Western blots, suggesting that the two antigens are similar or identical. Expression of EMA is a very early manifestation of neuronal differentiation; its distribution on growth cones suggests a role in mediating the interactions between growth cones and the external cues that guide them.

A variety of mechanisms have been proposed to explain the forward extension of cytoplasm in advancing cells and axonal growth cones, including actin polymerization and osmotic swelling. Based on our observations of the filopodia of cultured neuronal growth cones, we propose a mechanism involving motor-induced extension and retraction. We observed that filopodia (actin-based protrusions 0.2-0.5 mu in diameter) extend and retract from growth cone lamellae at the same rate. Further, force is generated at the tips of filopodia which is sufficient to produce compressive buckling of the proximal portion of the filopodium. From our analysis of these movements we suggest that a motor protein powers both the extension and retraction of filopodia.

Formation of the nervous system requires that neuronal growth cones follow specific paths and then stop at recognition signals, sensed at the growth cone's leading edge. We used antibody-coated gold particles viewed by video-enhanced differential interference contrast microscopy to observe the distribution and movement of two cell surface molecules, N-CAM and the 2A1 antigen, on growth cones of cultured cortical neurons. Gold particles are occasionally transported forward at 1-2 microns/s to the leading edge where they are trapped but continue to move. Concentration at the edge persists after cytochalasin D treatment or ATP depletion, but active movements to and along edges cease. We also observed a novel outward movement of small cytoplasmic aggregates at 1.8 microns/s in filopodia. We suggest that active forward transport and trapping involve reversible attachment of antigens to and transport along cytoskeletal elements localized to edges of growth cones.

To analyze cell lineage in the murine cerebral cortex, we infected progenitor cells with a recombinant retrovirus, then used the retroviral gene product to identify the descendants of infected cells. Cortices were infected on E12-E14 either in vivo or following dissociation and culture. In both cases, nearly all clones contained either neurons or glia, but not both. Thus, neuronal and glial lineages appear to diverge early in cortical development. To analyze the distribution of clonally related cells in vivo, clonal boundaries were reconstructed from serial sections. Perinatally (E18-PN0), clonally related cells were radially arrayed as they migrated to the cortical plate. Thus, clonal cohorts traverse a similar radial path. Following migration (PN7-PN23), neuronal clones generally remained radially arrayed, while glial clones were variable in orientation, suggesting that these two cell types accumulate in different ways. Neuronal clones sometimes spanned the full thickness of the cortex. Thus, a single progenitor can contribute neurons to several laminae.

Receptive fields of retinal ganglion cells were analyzed during extracellular microelectrode recordings in the optic tract of the lightly anesthetized pigeon. Four major types of receptive field can be distinguished among the 359 fibers studied. Twenty-five percent of the receptive fields are relatively simple, responding at on and at off to stationary spots of light in the central region. All of the receptive fields have inhibitory surrounds of varying strength that do not produce a response when illuminated alone, but antagonize responses from the central region. Motion sensitive units comprise 15% of the recorded population; they are similar to the on-off center type except that responses to stationary stimuli are absent or very weak while responses to moving stimuli are virorous. Directionally selective units also have the basic features of on-off, inhibitory surround cells, but respond to moving stimuli well from the preferred direction and not at all from the null direction. Directional cells have a broad range of null directions; in about one-third of the units the range becomes broader when the stimulus involves both center and surround of the receptive field, thus enhancing directional selectivity. Directionally selective units are common, comprising 38% of the units studied. Cells unresponsive to stimuli moving from anterior in the visual field are much more common than other types, while cells unresponsive to stimuli from posterior in the field are rare. A few units (11%) respond only at on or at off to stationary stimuli in their receptive field centers; they also have antagonistic but unresponsive receptive field surrounds. The area of the visual field sampled is uniform in regard to the relative numbers of the four major receptive field types.

The receptive-field properties of neurons in the striate visual cortex of normal and reeler mutant mice were studied with single-unit recording methods in order to determine whether the connections underlying these properties are altered by the developmental abnormality in neuronal position that characterizes reeler neocortex. Neurons with a projection through the corpus callosum were selected for study because they form a physiologically identifiable class of visual cortical neurons with a characteristic distribution of receptive-field properties that can be compared for normal and reeler cortex. Transcallosal cortical neurons in area 17 near its border with area 18a were identified by antidromic stimulation delivered through bipolar electrodes in the contralateral cortex. A computer controlled the visual stimuli, data acquisition, and analysis. Transcallosal neurons were principally found in layers II-III and V in the normal cortex and in a broand band deep in the reeler cortex. These populations had similar distributions of antidromic latencies, indicating that the neurons sampled from normal and reeler cortex were taken from populations with similar axonal diameters and soma sizes. The receptive-field properties of 46 units in 22 normal mice and 28 units in 11 reeler mice were characterized. Transcallosal neurons in both normal and reeler cortex were usually binocularly responsive and dominated by input from the contralateral eye. They exhibited either nonoriented (31 and 48%, respectively) or oriented (69 and 52%) receptive fields. Tuning 10 stimulus velocity was broad, with peak velocity sensitivities ranging from 1 to 1,000 degrees/s. Directional selectivity was present in 41% of normal units ad 32% of reeler units. There was no significant difference between normal and reeler cortex in the distribution of these properties. Transcallosal neurons were also examined for the presence of an inhibitory surround by comparing their responses to moving or stationary stimuli of varying sizes. Of the tested neurons, most (11/17 in normal cortex, 6/9 in reeler) showed evidence of a decrease in response to large moving stimuli. A large proportion (16/20) of normal neurons tested with stationary flashing stimuli had some degree of surround inhibition whereas significantly fewer (5/17) neurons in reeler cortex had this property. Thus, transcallosal neurons in reeler cortex less frequently had an inhibitory surround demonstrable with stationary flashing stimuli, but this difference between normal and reeler was not apparent with a moving stimulus.(ABSTRACT TRUNCATED AT 400 WORDS)

Efferents to the retina in the bird arise in the isthmo-optic nucleus of the caudal midbrain, and terminate on amacrine cells in the retina. The functional role of these efferents was studied by determining the receptive field properties of 107 optic tract fibers in the lightly anesthetized adult pigeon, and quantitating their responses to specific moving stimuli. While the recording from these fibers continued, the isthmo-optic nucleus was cooled by a thermoelectric cooling probe, and the response properties of the cells redetermined. Recording was maintained in half of the units long enough to observe recovery from cooling, and in several units the entire procedure was repeated. In 77 of the 107 units, responsiveness to all stimuli was decreased by removing efferent influences, whereas specific receptive field properties such as motion sensitivity or directionaltiy were not altered. All of the major receptive field types were affected in a similar fashion, irrespective of their position in the visual field. Responses to stimuli that did not involve the antagonistic surround were similarly affected by removal of the efferents, as were units were both weak and strong antagonistic surrounds. Efferents exert their influence on retinal ganglion cells by way of the amacrine cells on which they terminate. Data available on amacrines in Necturus indicates that they are inhibitory to ganglion cells. If amacrines have a similar role in the pigeon, then it may be stated that decreased activity in the centrifugal fibers leads to enhanced inhibition throughout the receptive fields of ganglion cells, and increased activity in the efferents produces disinhibition.

We studied the receptive field properties of single neurons in the primary visual cortex (area 17) of the mouse and the distribution of receptive field types among the cortical laminae. Three basic receptive field types were found: 1) Cells with oriented receptive fields, many of which could be classified as simple or complex, were found in all layers of the cortex, but occurred with greater frequency in layers II and III and less commonly in Layer IV. 2) Cells with non-oriented receptive fields had ON, OFF, or ON-OFF centers; they were found in all layers but were predominant in layer IV. Two subclasses of non-oriented receptive fields were characterized based on their responses to stationary and moving stimuli. One group of cells with non-oriented receptive fields responded vigorously with sustained firing to stationary flashing stimuli, and also responded well to moving stimuli over a wide range of stimulus velocities. A second group of non-oriented cells, termed motion-selective, responded poorly or not at all to stationary stimuli and responded optimally to moving stimuli over a restricted range of velocities. 3) A distinct group of neurons, termed large field, non-oriented (LFNO) cells, were found almost exclusively in layer V. LFNO cells had receptive fields that were larger than those of the other two major classes at all visual-field locations; they also had higher rates of spontaneous activity and responded to higher stimulus velocities than the other classes. In these respects, LFNO cells resembled the layer V cells of area 17 in the cat and the layer V and VI cells of area 17 in the monkey that project to the superior colliculus. We injected horseradish peroxidase into the superior colliculus, and determined that corticotectal cells in the mouse were also located in layer V, the layer where we recorded LFNO cells. Additional evidence that some LFNO cells project to the superior colliculus was provided by preliminary experiments in which we stimulated the superior colliculus and antidromically activated cortical cells with LFNO receptive fields. Neurons with LFNO receptive fields thus constitute a class that is functionally distinct, with cell bodies that are located in a single layer (V) of area 17 in the mouse.

In contrast to the traditional view that striate visual cortex (area 17) is surrounded by two homogeneous cortical areas (areas 18 and 19), recent studies have shown that mammalian extrastriate visual cortex contains several anatomically and functionally distinct subregions. One such region, the V-4 complex of the rhesus monkey, is highly specialized for the analysis of color information, suggesting that a lesion in a homologous region might produce a defect in color vision while sparing other visual functions. We have studied a patient whose clinical syndrome supports this suggestion: a 44-year-old man with normal color vision suffered two cerebral infarctions that produced first a right and then a left superior homonymous quadrantanopia and also caused prosopagnosia, topographical disorientation, and severely impaired color vision. Computed tomography demonstrated extensive lesions in both inferior occipital lobes in the territories of the lateral branches of the posterior cerebral arteries, involving the lingual and medial occipitotemporal gyri bilaterally; these gyri contain the inferior portion of striate cortex and segments of extrastriate visual cortex. The patient had no difficulty in giving the correct color names associated with common objects presented either verbally or in outline drawings. Standardized testing with the Farnsworth-Munsell 100-hue test, the Nagel anomaloscope, and a method that tests for just-noticeable differences between monochromatic stimuli all showed that the patient's ability to distinguish one color from another was markedly imparied but not totally absent. In contrast, visual acuity, reading, visually guided eye movements, and stereopsis were normal. Cells in the V-4 complex of monkey extrastriate cortex are highly specialized for distinguishing one color from another; the hue discrimination deficit that was demonstrated in this patient with cerebral color blindness indicates that a region or regions with similar function has been damaged.

A conceptual framework is proposed for the selection of monitored parameters during anesthesia, and a new device for monitoring the parameters in a unified manner is briefly presented. A 'basic set' of 6 parameters is proposed to cover the needs of most routine anesthesia: Blood Pressure, ECG/Heart Rate, Temperature, FiO2, FetCO2, and, Cortical Activity (by EEG spectral analysis). Additional parameters are added in accordance with specified factors such as patient status and complexity of the surgical procedure. An initial version of a new monitor, 'Cerebro Trac', designed for neurosurgery and cardiovascular surgery, is briefly presented, along with planned future capabilities and directions for its use.

In order to analyze the role of lamination in establishing the precisely ordered connectional pattern of the neocortex, we compared the afferent and efferent connections of the visual cortical areas in normal mice with those of the mutant mouse reeler (rl). The reeler mutation causes disruption of the laminar organization of the neocortex; all classes of neurons are present but are abnormally located. The corticocortical and thalamocortical connection os visual cortical areas 17, 18a, and 18b were determined in normal and reeler mice with injections of horseradish peroxidase (HRP) or HRP conjugated with wheat germ agglutinin (HRP-WGA). The diffusion of HRP-WGA is highly restricted due to the surface binding properties of the lectin; it was particularly effective in demonstrating retinotopically ordered connections. We found that the patterns of connections made the reeler mutant are indistinguishable from normal. Cortical loci in area 17 are reciprocally connected to homotopic locations in areas 18a and 18b. Area 17 is also reciprocally connected with dorsal lateral geniculate nucleus of the thalamus and projects to the superior colliculus. Areas 18a and 18b are reciprocally connected with each other and with the lateral posterior and lateral nuclei of the thalamus, respectively. In addition, we found evidence of reciprocal connections between the lateral posterior nucleus and area 17, and between the lateral nucleus and areas 17 and 18a. The results indicate the neurons in visual cortical areas of the reeler mutant mouse are capable of forming retinotopically organized corticocortical and thalamocortical connections in a pattern similar to that found in normal animals. Thus the genetic anomaly producing incorrect neuronal positioning during development of the reeler cortex does not seriously impede the pathway and target recognition mechanisms responsible for formation of functionally appropriate cortical connections.

Detailed retinotopic maps of primary visual cortex (area 17) and the extrastriate visual regions surrounding it (areas 18a and 18b) have been constructed for the C57BL/6J mouse using standard electrophysiological mapping techniques. Primary visual cortex (area 17), as defined cytoarchitectonically, contains one complete representation of the contralateral visual field, termed V1, in which azimuth and elevation lines are approximately orthogonal. The upper visual field is represented caudally and the nasal field laterally. Binocular cells are encountered in the cortical representation of the nasal 30--40 degrees of the visual field, and there is an expanded representation of the nasal field. Extrastriate visual cortex of the mouse, like that of other mammals, contains multiple representations of the visual field. The cytoarchitectonic region of cortex lateral and rostral to area 17, termed area 18a, contains at least two such representations. The more medial of these, which by convention we have called V2, is a narrow strip surrounding V1 on its lateral and rostral aspects; the vertical meridian lies along a portion of its common border with V1. The visual field representation in V2 is not a mirror image of that in V1; the representation of the horizontal meridian forms the lateral border of V2, and the visual field representation is split so that adjacent points on either side of the horizontal meridian are represented in nonadjacent parts of V2. The other visual field representation within area 18a, which we have termed V3, is a small but apparently complete representation that lies lateral to V2. The visual field representations medial to area 17 correspond to cytoarchitectonic area 18b. Area 18b contains two representations of the temporal visual field that we have labeled Vm-r and Vm-c, and contains little or no representation of the most nasal aspect of the field.

1. Peripheral mechanisms that might contribute to colour vision in the cat have been investigated by recording from single units in the lateral geniculate and optic tract. Evidence is presented that the input to these cells comes from a single class of cones with a single spectral sensitivity.2. In cats with pupils dilated a background level of 10-30 cd/m(2) was sufficient to saturate the rod system for all units. When the rods were saturated, the spectral sensitivity of all units peaked at 556 nm; this was true both for centre and periphery of the receptive field. The spectral sensitivity curve was slightly narrower than the Dartnall nomogram. It could not be shifted by chromatic adaptation with red, green, blue or yellow backgrounds.3. In the mesopic range (0.1-10 cd/m(2)), the threshold could be predicted in terms of two mechanisms, a cone mechanism with spectral sensitivity peaking at 556 nm, and a rod mechanism with spectral sensitivity at 500 nm. The mechanisms were separated and their increment threshold curves measured by testing with one colour against a background of another colour. All units had input from both rods and cones. The changeover from rods to cones occurred at the same level of adaptation in both centre and periphery of the receptive field. Threshold for the cones was between 0.04 and 0.25 cd/m(2) with the pupil dilated, for a spot covering the centre of the receptive field.4. None of the results was found to vary between lateral geniculate and optic tract, with layer in the lateral geniculate, or with distance from area centralis in the visual field.5. The lack of evidence for more than one cone type suggests that colour discrimination in the cat may be a phenomenon of mesopic vision, based on differences in spectral sensitivity of the rods and a single class of cones.

The migration of the screening pigment was investigated in the retina of the intact squid. The action spectrum of pigment migration corresponds to the action spectrum of the visual pigment, rhodopsin, rather than to the absorption spectrum of the screening pigment. The total number of quanta required for a fixed criterion of pigment migration is the same, when the quanta are delivered over any period of time from 6 s to an hour or more. When less than 3-10% of the rhodopsin is isomerized, the screening pigment migrates out to the tips of the receptors with a time-course of 5-15 min, and back again over the same period of time. When rather more than 10% is isomerized, the outward migration takes 5-15 min, but the screening pigment does not migrate inwards, even after several hours in the dark. Indirect evidence suggests that the band of screening pigment, when it reaches the tips of the receptors, is approximately equivalent to a filter of 0.6 log units. The spectral sensitivity of the optic nerve response was measured, and was found to be broader than the absorption spectrum of squid rhodopsin in vitro; the broadness could be explained by self-screening, assuming a density of rhodopsin of 0.6 log units at 500 nm.

1. The ability of cats to distinguish colours was investigated at mesopic and photopic levels to test the hypothesis that cats discriminate wavelength by using rods in conjunction with a single type of cone.2. Cats were trained to distinguish red from cyan, and orange from cyan at the mesopic level. They retained the ability to make this discrimination when the coloured stimuli were placed against a background bright enough to saturate the rods.3. One cat was also tested after being exposed to a bright white light of 9000 cd/m(2) for a period of 5 min, and found able to distinguish red from cyan.4. These results suggest that cats have more than one type of cone. Subsequent recordings from single units in the lateral geniculate nucleus showed that there are rare opponent colour units in layer B with input from a green-absorbing cone and a blue-absorbing cone.

A microelectrode survey of the cat lateral geniculate has uncovered an infrequent new type of lateral geniculate cell in layer B with "on" center responses to short wavelengths and "off" center responses to long wavelengths. The short wavelength responses are mediated by cones with peak sensitivity at about 450 nanometers, and the long wavelength responses by cones with peak sensitivity at 556 nanometers. Two of double opponent color cells also had double opponent features.