Effects of bicuculline on signal detectability in lateral geniculate nucleus relay cells

The lateral geniculate nucleus conveys the center-surround organized retinal receptive fields to the cortex in a way that does not significantly alter their spatial structure. However, non-retinal influences may change the 'strength' (detectability) of the signal under conditions of anesthesia, arousal and attention. A previous analysis of receiver operating characteristic curves in cat suggests that a reduction in signal detectability occurs in lateral geniculate nucleus (LGN) relay cells in anesthetized animals in comparison to the retinal afferents. In the present study, it was found that antagonism of GABAA receptors with bicuculline (BIC) increased signal detectability in LGN relay cells in the tree shrew (Tupaia belangeri). This change is consistent with the hypothesis that feedforward and/or feedback GABAergic circuits in the LGN differentially affect the retinogeniculate transfer ratio for visually driven activity versus maintained (spontaneous) activity. Under conditions of arousal or attention, signal detectability may be increased by brainstem activation, thus increasing the flow of information in the visual system.     

Effects of bicuculline on direction-sensitive relay cells in the dorsal lateral geniculate nucleus (LGNd) of cats

The direction sensitivity of relay cells in the cat's dorsal lateral geniculate (LGNd) was measured using sinusoidal grating stimuli before and during local bicuculline administration. One hundred and twenty-eight LGNd relay cells were recorded in laminae A and A1, of which 44 relay cells (34%) were found to be sensitive to direction of stimulus movement. The direction-sensitive LGNd relay cells could be differentiated into two subgroups based on different measures of their response amplitude. Type I cells exhibited their direction sensitivity when the fundamental Fourier component (FFC) of the poststimulus time histograms (PSTHs) was used as response measure, but did not show significant direction sensitivity when mean firing rate was used. Type II cells exhibited their direction sensitivity, no matter whether the FFC or mean firing rate was used as the measure. Of 35 cells analyzed, 27 cells remained direction sensitive during bicuculline administration. At the population level, the direction bias of type I cells did not change systematically, while the direction bias of type II cells decreased significantly during bicuculline administration. These results suggest that the direction bias of these two types of relay cells are mediated by different neural mechanisms. The direction bias of type I cells may involve multiple inputs from spatio-temporally separate subunits within retinal ganglion cells receptive fields. The direction bias of type II cells may involve GABAergic neuronal circuits within the LGNd.     

Organized arrangement of orientation-sensitive relay cells in the cat's dorsal lateral geniculate nucleus

We studied the physiological orientation biases of over 700 relay cells in the cat's dorsal lateral geniculate nucleus (LGNd). Relay cells were sampled at regular intervals along horizontally as well as vertically oriented electrode penetrations in a fashion analogous to that used previously in studies of visual cortex (Hubel and Wiesel, 1962). The strengths of the orientation biases and the distributions of the preferred orientations were determined for different classes of relay cells, relay cells in different layers of the LGNd, and relay cells subserving different parts of the visual field. We find that, at the population level, LGNd cells exhibit about the same degree of orientation bias as do the retinal ganglion cells providing their inputs (see also Soodak et al., 1987). Also, as in the retina (Levick and Thibos, 1982; Leventhal and Schall, 1983), most LGNd cells tend to prefer stimuli oriented radially, i.e., parallel to the line connecting their receptive fields to the area centralis projection. However, the radial bias in the LGNd is weaker than in the retina. Moreover, there is a relative overrepresentation of cells preferring tangentially oriented stimuli in the LGNd but not in the retina. As a result of the overrepresentation of cells preferring radial and tangential stimuli, the overall distribution of preferred orientations varies in regions of the LGNd subserving different parts of the visual field. Reconstructions of our electrode penetrations provide evidence that, unlike in the retina, cells having similar preferred orientations are clustered in the LGNd. This clustering is apparent for all cell types and in all parts of laminae A and A1. The tendency to cluster according to preferred orientation is evident for cells preferring radially, intermediately, and tangentially oriented stimuli and thus is not simply a reflection of the radial bias evident among retinal ganglion cells at the population level. It is already known that cells having inputs from different eyes, on-center, off-center, X-, Y-, W-type, and color-sensitive ganglion cells are distributed nonrandomly in the LGNd of cats and monkeys (for review, see Rodieck, 1979; Stone et al., 1979; Lennie, 1981; Stone, 1983). The finding that relay cells having similar preferred orientations are also distributed nonrandomly suggests that the initial sorting of virtually all properties segregated in visual cortex may begin in the LGNd.     

Effects of bicuculline-induced epileptiform activity on development of receptive field properties in striate cortex and lateral geniculate nucleus of the rabbit

Studies have shown that normal development of receptive field properties in striate cortex and lateral geniculate nucleus (LGN) of the rabbit is severely altered in the presence of penicillin-induced disruption of cortical neuronal activity. We wished to replicate these studies using a different convulsant drug in order to rule out possible effects due to the penicillin drug itself. Aqueous bicuculline was injected twice daily into a cannula implanted over the monocular region of one striate cortex. Drug administration was initiated on postnatal day 8-9 and was discontinued either on postnatal day 19-24 or on postnatal day 24-30 for studies of the LGN and striate cortex, respectively. Coincidental with the bicuculline injections, control solutions were similarly applied to the monocular region of the contralateral striate cortex. Single-unit recordings made from LGN ipsilateral to bicuculline-treated cortex revealed normal percentages of receptive field types. However, in single-unit recordings made from bicuculline-treated striate cortex, an abnormal percentage distribution of receptive field types was found. In such cortex there was an unusually high proportion of no-response type cells and a substantially reduced proportion of oriented type cells. These developmental abnormalities are virtually the same as those found in the striate cortex of similarly reared animals treated with penicillin. Our present results lend support to our previous conclusion that in the rabbit, disruption of orderly neuronal activity in the geniculostriate system has a detrimental effect on the development of receptive fields in this system.     

Modulatory effects mediated by metabotropic glutamate receptor 5 on lateral geniculate nucleus relay cells

Glutamate is thought to be the excitatory neurotransmitter in the lateral geniculate nucleus (LGN) of the cat, mediating visual transmission from the retina via ionotropic receptors of both D,L-alpha-amino-3-hydroxy-5-alpha-methyl-4-isoxazolepropionate and N-methyl-D-aspartate subtypes. Moreover, glutamate also exerts an important modulatory influence on LGN cells, where metabotropic glutamate receptors (mGluRs) seem to play a crucial role. Here we show in anesthetized adult cats that iontophoretic application of the specific mGluR5 antagonist 2-methyl-6-(phenylethynyl)-pyridine (MPEP) produced two, distinctly different, effects on LGN neurons. Visual responses to flashing spots and drifting gratings were attenuated (decreased by an average of 59%) in 13 of 23 of the cells but augmented (increased by an average of 60%) in 10 of 23 of the cells. Further, in each case when the specific mGluR5 agonist (R,S)-2-chloro-5-hydroxyphenylglycine was applied, the effects obtained were the opposite to those of MPEP. Data obtained in a second group of experiments to determine a possible interaction between mGluR5 blockade by MPEP and glutamate ionotropic receptors show that, in the majority of neurons (11 of 15, 73%), the MPEP-mediated effects seem to be independent of N-methyl-D-aspartate and D,L-alpha-amino-3-hydroxy-5-alpha-methyl-4-isoxazolepropionate receptor activity. Our results demonstrate a physiological role for mGluR5 in controlling retinal input and show, in vivo, a more intricate scenario than previously suggested, highlighting the complexity of metabotropic receptor interactions with excitatory and inhibitory elements in the thalamus.     

Influence of the visual cortex upon receptive field organization of lateral geniculate cells in rabbits

In anesthetized rabbits, the receptive fields of lateral geniculate cells were mapped prior to and following the interruption of the corticogeniculate feed-back. Visual cortex (V.C.) was depressed by a focal application of 3 M KCl. The responsiveness of the V.C. was verified by monitoring the visually evoked potentials. In off- and on-center cells, the surround excitatory responses were remarkably reduced and even fully abolished in most units. In contrast, the center excitation remained unmodified. These effects were reversible. In some on-center units the center response had also decreased, and was replaced by an evoked inhibitory response. Relay cells and interneurons which yielded on and off responses over the entire area of the receptive field exhibited a loss of only one of the evoked discharges. It is concluded that the V.C. exerts mostly a specific desinhibitory action upon the geniculate network. This action affects either the center or the surround responses differentially. The results are compared with those obtained from cats.     

Temporal properties of colour opponent receptive fields in the cat lateral geniculate nucleus

The primordial form of mammalian colour vision relies on opponent interactions between inputs from just two cone types, ‘blue’ (S‐) and ‘green’ (ML‐) cones. We recently described the spatial receptive field structure of colour opponent blue‐ON cells from the lateral geniculate nucleus of cats. Functional inputs from the opponent cone types were spatially coextensive and equally weighted, supporting their high chromatic and low achromatic sensitivity. Here, we studied relative cone weights, temporal frequency tuning and visual latency of cat blue‐ON cells and non‐opponent achromatic cells to temporally modulated cone‐isolating and achromatic stimuli. We confirmed that blue‐ON cells receive equally weighted antagonistic inputs from S‐ and ML‐cones whereas achromatic cells receive exclusive ML‐cone input. The temporal frequency tuning curves of S‐ and ML‐cone inputs to blue‐ON cells were tightly correlated between 1 and 48 Hz. Optimal temporal frequencies of blue‐ON cells were around 3 Hz, whereas the frequency optimum of achromatic cells was close to 10 Hz. Most blue‐ON cells showed negligible response to achromatic flicker across all frequencies tested. Latency to visual stimulation was significantly greater in blue‐ON than in achromatic cells. The S‐ and ML‐cone responses of blue‐ON cells had on average, similar latencies to each other. Altogether, cat blue‐ON cells showed remarkable balance of opponent cone inputs. Our results also confirm similarities to primate blue‐ON cells suggesting that colour vision in mammals evolved on the basis of a sluggish pathway that is optimized for chromatic sensitivity at a wide range of spatial and temporal frequencies.     

Morphological organization of thalamocortical relay cells in the dorsal lateral geniculate nucleus of the North American opossum

The morphology of relay cells in the dorsal lateral geniculate nucleus of the North American opossum was studied by using both Golgi-Cox material and cells stained from retrograde transport of horseradish peroxidase. In general, soma sizes were largest in the part of the nucleus representing the central retina and decreased from the middle third of the nucleus to the anterior to posterior poles. Relay cells labeled with horseradish peroxidase were found to constitute approximately 90% of the dorsal lateral geniculate nucleus cells and have larger soma diameters than most unlabeled cells. From morphometric analysis of several structural characteristics, three classes of relay cells were identified in both Golgi-Cox and horseradish peroxidase material. Type 1 cells, the predominant class, exhibited radially arranged primary dendritic fields, symmetrically organized relative to projection lines. Type 2 cells had relatively few primary dendrites, and complex dendritic fields that were oriented parallel to projection lines. Least numerous were Type 3 cells, which were characterized by relatively sparse dendritic fields oriented perpendicular to projection lines. An additional class of neuron, Type 4 cells, with small somata and sparse dendritic branching, was found only in Golgi-Cox material. Cells with Type 4 dendritic morphology were not found with retrograde horseradish peroxidase labeling and may represent interneurons. The classification of morphologically characterized cells in the opossum dorsal lateral geniculate nucleus was evaluated quantitatively with multivariate discriminant analysis. The classes are compared to physiologically identified Y-, X-, and W-like relay cells in the opossum and to relay cell classes in other species.     

Ventral lateral geniculate nucleus projects to the dorsal lateral geniculate nucleus in the cat

The lamina C3 of the dorsal lateral geniculate nucleus of the cat does not receive retinal projections but instead receives visual information from the small subpopulation of W-type ganglion cells via the upper substratum of the stratum griseum superficiale of the superior colliculus. We herein report a projection from the lateral division of the ventral lateral geniculate nucleus into the lamina C3 of the dorsal lateral geniculate nucleus. As the lateral division receives projections from the contralateral retina and the ipsilateral upper stratum griseum superficiale of the superior colliculus, we suggest that these regions make up a small cell type W-cell neuronal network that provides visual information to layer I of the striate cortex via the lamina C3.     

Laminar organization of receptive field properties in the dorsal lateral geniculate nucleus of the tree shrew (Tupaiaglis belangeri)

A laminar analysis of the receptive field properties of relay cells in the binocular region of the tree shrew dorsal lateral geniculate nucleus (LGN) found three main subdivisions. Lamina 1 (receiving ipsilateral eye input) and lamina 2 (contralateral) comprise a pair of layers that contain only ON-center neurons. Laminae 4 (contralateral) and 5 (ipsilateral) comprise a pair of layers with mostly OFF-center cells (86%). Laminae 3 and 6 (both contralaterally innervated) also form a distinct pair, although lamina 3 contains a mixture of cells with ON-centers (43%) or OFF-centers (57%), and lamina 6 contains mostly cells with ON-OFF centers and suppressive surrounds (81%). Cells located in the interlaminar zones resembled neurons in laminae 3 and 6. In comparison with the cells in the OFF-center laminae 4 and 5, the ON-center cells in laminae 1 and 2 had smaller, more elliptical receptive field centers with stronger responses to flashed visual stimuli. In addition, cells in the ipsilateral eye laminae 1 and 5 showed a greater change in center diameter, with eccentricity from the area centralis, than cells in the contralateral eye laminae 2 and 4. Principal components analysis using six receptive field properties (latency to optic chiasm stimulation, receptive field center diameter, maintained discharge rate, response onset latency, peak spike density, and phasic-tonic index) suggested that the cells in laminae 3 and 6 and the interlaminar zones are W-like. Principal components analysis of the same receptive field properties in laminae 1, 2, 4, and 5 did not reveal differences clearly related to X-like (parvocellular) and Y-like (magnocellular) categories. Ninety-seven percent of the cells tested for linearity of spatial summation in laminae 1, 2, 4, and 5 were linear. We conclude that the dominant organizational features of the tree shrew LGN are the ON-center, OFF-mter, and W pairs of layers that project to different regions within the striate cortex. © 1995 Wiley-Liss, Inc.     

Morphologically distinct classes of relay cells exhibit regional preferences in the dorsal lateral geniculate nucleus of the mouse

A fundamental feature of the mammalian visual system is the presence of separate channels that work in parallel to efficiently extract and analyze specific elements of a visual scene. Despite the extensive use of the mouse as a model system, it is not clear whether such parallel organization extends beyond the retina to subcortical structures, such as the dorsal lateral geniculate (dLGN) of thalamus. To begin to address this, we examined the morphology of biocytin-filled relay cells recorded in dLGN of mice. Based on a quantitative assessment of their dendritic architecture, we found that even at early postnatal ages relay cells could be readily classified as X-like (biconical), Y-like (symmetrical), or W-like (hemispheric) and that each cell type was regionally specified in dLGN. X-like cells were confined primarily to the monocular ventral region of dLGN. Y-like cells occupied a central core that also contained ipsilateral eye projections, whereas W-like cells were found along the perimeter of dLGN. Similar to cat, Y-like cells were more prevalent than X- and W-like cells, and X-like cells tended to be smaller than other cell types. However, the dendritic fields of X- and W-like cells did not exhibit an orientation bias with respect to optic tract or boundaries of dLGN. Although we found clear morphological differences among relay cells, an analysis of their electrophysiological properties did not reveal any additional distinguishing characteristics. Overall, these data coupled with recent observations in the retina suggest that the mouse has many of the hallmark features of a system-wide parallel organization.     

The influence of GABAergic inhibitory processes on the receptive field structure of X and Y cells in cat dorsal lateral geniculate nucleus (dLGN)

Visually elicited inhibitory processes, underlying the receptive field structure of cells in layers A and A1 of the cat dorsal lateral geniculate nucleus (dLGN), have been examined by a combination of visual neurophysiological and iontophoretic techniques. Discrete visual stimulation of both centre and surround mechanisms, produced a powerful suppression of the elevated background discharge levels induced by iontophoretic application of an excitatory amino acid. These observations are consistent with the activation of a postsynaptic inhibitory input, a view supported by the fact that the suppressive effects were blocked by iontophoretic application of bicuculline, an antagonist of GABA, a putative inhibitory transmitter in the dLGN. These inhibitory effects were always elicited by the opposite phase of a flashed stimulus to that eliciting responses associated with the receptive field region. That is 'on' inhibitory effects were elicited from 'off' excitatory regions and 'off' inhibitory effects from 'on' excitatory regions. Plotting the responses of dLGN cells to flashed stimuli of increasing diameter, before and during iontophoretic application of bicuculline, revealed a marked reduction of the surround antagonism of centre response in the presence of the drug. During bicuculline application, surround antagonism of centre responses was at a level associated with that seen in retinal ganglion cells. Annular stimuli of internal and external diameter selected to be just outside the centre-surround border, insofar as they gave pure surround responses, were observed in the presence of bicuculline to elicit strong centre responses, as well as surround responses. These observations indicate that GABAergic inhibitory processes generate the enhanced centre-surround antagonism associated with dLGN cells and serve to increase the contrast between the two sets of mechanisms at the centre-surround border. The present conclusions apply equally to 'X' and 'Y' cells.     

Effects of early monocular eyelid suture upon development of relay cell classes in the cat's lateral geniculate nucleus

Horseradish peroxidase (HRP) was injected into visual cortex of four normal cats and five cats raised with monocular lid suture, and retrograde labelling was assessed in cells of the lateral geniculate nucleus. In all but one of the sutured cats (noted below) focal injections were carefully limited to area 17 or 18 and analysis of labelling focused on laminae A and A1. The effects of deprivation were indistinguishable whether lamina A or A1 was deprived, and in all cases, the nondeprived laminae had labelling essentially identical to that seen in normal cats. After area 17 injections (bilateral in one normal cat and unilateral in 3 deprived cats), roughly 77% of the cells in nondeprived laminae were labelled and they were mostly small to medium in size. Deprived laminae, when compared to nondeprived laminae, had two abnormalities: (1) cells, both labelled and unlabelled, were smaller; and (2) roughly 11% fewer cells (i.e., 66%) were labelled, and this represents a small but statistically significant difference for each cat. After area 18 injections (bilateral in one normal cat plus unilateral in 3 other normal and 3 deprived cats), roughly 15% of the cells in nondeprived laminae were labelled, and they tended to be large in size. Deprived laminae, when compared to nondeprived laminae, had three abnormalities: (1) only 5–6% of the cells were labelled, and these tended to be quite faintly labelled; (2) the volume occupied by labelled cells was small; and (3) both labelled and unlabelled cells were reduced in size. Finally, large bilateral injections were made throughout occipitotemporal cortex in one lid sutured cat in an effort to label completely the terminal zones of cells in the medial interlaminar nucleus (MIN), a division of the lateral geniculate nucleus; this cat also had a prior intraocular injection of tritiated proline to provide through subsequent autoradiography a delineation of deprived and nondeprived portions of MIN. Roughly 78% of the cells in nondeprived portions of MIN were labelled in this cat. In the deprived portions, only about 51% of the cells were labelled, and these tended to be faintly labelled. Also, labelled cells were smaller, and unlabelled cells were larger in deprived than they were in nondeprived portions. Since prior studies have shown that, within the A laminae, X-cells project exclusively to area 17 whereas the Y-cell population projects to areas 17 and 18, these data are taken as further support of the conclusion that geniculate Y-cells are more seriously affected by the early deprivation than are geniculate X-cell. That is, these data are consistent with the suggestion that a similar population of Y-cells in deprived laminae (roughly 10% of the overall cell total) fail to transport HRP from area 17 or area 18 injections. This can be extended to the MIN, which seems to be comprised nearly exclusively of Y-cells. However, these conclusions must be considered tentative, since interpretation of HRP data can be difficult as evidenced by discrepancies in the literature.     

The visual field representation of the rat ventral lateral geniculate nucleus

The representation of the visual field in the ventral lateral geniculate nucleus (LGNv) was studied in rats anesthetized with urethane by recording the response of single units to visual stimulation. Receptive fields of LGNv units were plotted on a campimeter, 60 cm in diameter, which was placed 30 cm from the contralateral eye. LGNv neurons responded mainly to stimulation of the contralateral eye with on-tonic characteristics. Few neurons responded only to stimulation of the ipsilateral eye and no binocular interaction was observed. Retinotopic organization was clearly seen in the LGNv; the nasal visual fields were represented dorsally, the temporal fields ventrally, and the upper to lower visual fields were in the rostrolateral to caudomedial parts of the LGNv. A given point in the visual field is represented along a line running through the LGNv in a rostrocaudal direction. Almost the entire horizontal extent of the contralateral visual field was represented in the LGNv, whereas vertically the visual field between 40 degrees above and 20 degrees below the distribution axis was represented. The major axis of the strip of the visual field containing all the RF centers, which is referred to as the distribution axis, inclined nasally up and temporally down at an angle of 10.4 degrees to the 0 degree horizontal meridian line. The representation of the distribution axis in the retina was in accordance with the major axis of retinal ganglion cell distribution (Fukuda, '77; Schober and Gruschka, '77).     

The representation of the visual field in the lateral geniculate nucleus of Macaca mulatta

Microelectrode recording techniques were used to investigate the projection of the visual field into the lateral geniculate nucleus (LGN) of Macaca mulatta. The data were used to construct charts plotting visual direction, designated in terms of azimuth and elevation, onto sections of the nucleus cut in coronal, sagittal and horizontal Horsley-Clarke planes. The projection of the horizontal meridian divides the LGN along its plane of symmetry into a medial-superior half having negative elevations and a lateral-inferior half having positive elevations. Elevations become more positive or negative with distance from this plane. Azimuths closest to the vertical meridian are located posteriorly, while the most peripheral azimuths are found at the anterior pole. Two families of surfaces representing visual directions of constant azimuth and elevation are described. Visual field zones of increasing eccentricity are represented serially along the posterior-anterior axis of the LGN, with the foveal area restricted to the posterior pole and the monocular crescent projecting to the anterior pole. The mapping is completely continuous across the horizontal meridian. The edges of the stacked cell laminae exposed around the periphery of the LGN form an oval band which receives the projection of the perimeter of the contralateral hemifield. The vertical meridian is represented by the posterior two-thirds of this band, while the periphery of the hemifield projects to the anterior third. The central visual field out to the optic disc is represented by six cell layers, while the rest of the binocular field projects to four layers only (2 parvocellular and 2 magnocellular). The monocular crescent is represented by one parvocellular and one magnocellular layer. Features associated with the projection column of the optic disc are integrated into the transition from six to four layers. Details of the receptive field topography in the vicinity of the optic disc discontinuities indicate that these gaps are produced by intralaminar mechanisms. The magnification factor (mm³/steradian) increases monotonically from peripheral visual fields to the foveal center, varying over a range of three decades. This range is intermediate between those derived from data reported in the literature for the retina and the striate cortex. The ratio of LGN magnifications at any two angular eccentricities is a power function, with an exponent of 1.34, of the corresponding ratio of retinal ganglion cell densities. Similarly, the ratio of cortical magnifications (mm²/steradian) at any two eccentricites is a power function, with an exponent of 1.35, of the corresponding ratio of LGN magnifications.     

The identification of relay neurons in the dorsal lateral geniculate nucleus of monkeys using horseradish peroxidase.

Intraaxonal retrograde transport of the protein horseradish peroxidase (HRP) was used to identify relay neurons in the dorsal lateral geniculate nucleus (LGN) of owl (Aotus trivirgatus) and rhesus (Macaca mulatta) monkeys. In both species, from 94.1-98.6% of the neurons within columns extending through both parvocellular and magnocellular layers were labeled following injection of HRP into striate cortex. Labeled neurons were also identified in the thin ventral-most S(0) Layers. Although most of the cells within the thin interlaminar regions in the LGN of both species were labeled following injections of HRP, many unlabeled neurons were identified within the large cell-rich interlaminar region (IL) between the internal parvocellular and internal magnocellular layers in the LGN of the owl monkey, suggesting that IL may be a specialized region containing a large number of intrinsic neurons. Finally, measurement of the cell diameters of neurons within the densely labeled areas in relay layers revealed that labeled and unlabeled neurons could not be distinguished on the basis of cell body size alone and that some of the smallest cells of the LGN project to striate cortex. These findings indicate that nearly all of the neurons of the main relay layers of the LGN in these two primates are relay cells and that the organization of the LGN in primates may differ significantly from that of other mammals with respect to the percentage of interneurons.     

Effects of monocular closure at different ages on deprived and undeprived cells in the primate lateral geniculate nucleus

This study has examined the effects of monocular visual deprivation on cells in the lateral geniculate nucleus of the primate by comparing the sizes of cells in deprived and undeprived LGN laminae of experimental rhesus monkeys with those of cells in the corresponding laminae of normal animals. A number of conclusions may be drawn from this comparison: monocular visual deprivation has major effects on cells in the undeprived LGN laminae and these vary with age at closure; the initial effect of monocular closure from birth is to cause marked hypertrophy of undeprived parvocellular cells with little shrinkage of the deprived parvocellular cells, whereas late monocular closure (after 2 months of age) causes marked shrinkage of both undeprived and deprived parvocellular cells; following monocular closure at birth, the LGN abnormality continues to evolve until at least 3 months of age, with a marked parallel shrinkage affecting both deprived and undeprived parvocellular cells. The initial hypertrophy of the undeprived cells is reversed and the deprived cells become smaller than normal; cells in the monkey LGN are sensitive to visual deprivation until about 1 year of age, much later than previously thought. Visual experience, however, modifies this sensitivity so that the effects of monocular visual deprivation are both qualitatively and quantitatively different at different ages; there are important differences between the susceptibility of cells in the magnocellular and parvocellular laminae to visual deprivation; and actual shrinkage of cells to markedly below normal size occurs and the smaller size is not simply failure of growth.