A note on the problem of retinal projections to the inferior pulvinar nucleus of primates

Retinal ganglion cells projecting to superior colliculus and pulvinar in marmoset

We determined the retinal ganglion cell types projecting to the medial subdivision of inferior pulvinar (PIm) and the superior colliculus (SC) in the common marmoset monkey, Callithrix jacchus. Adult marmosets received a bidirectional tracer cocktail into the PIm (conjugated to Alexa fluor 488), and the SC (conjugated to Alexa fluor 594) using an MRI-guided approach. One SC injection included the pretectum. The large majority of retrogradely labelled cells were obtained from SC injections, with only a small proportion obtained after PIm injections. Retrogradely labelled cells were injected intracellularly in vitro using lipophilic dyes (DiI, DiO). The SC and PIm both received input from a variety of ganglion cell types. Input to the PIm was dominated by broad thorny (41%), narrow thorny (24%) and large bistratified (25%) ganglion cells. Input to the SC was dominated by parasol (37%), broad thorny (24%) and narrow thorny (17%) cells. Midget ganglion cells (which make up the large majority of primate retinal ganglion cells) and small bistratified (blue-ON/yellow OFF) cells were never observed to project to SC or PIm. Small numbers of other wide-field ganglion cell types were also encountered. Giant sparse (presumed melanopsin-expressing) cells were only seen following the tracer injection which included the pretectum. We note that despite the location of pulvinar complex in dorsal thalamus, and its increased size and functional importance in primate evolution, the retinal projections to pulvinar have more in common with SC projections than they do with projections to the dorsal lateral geniculate nucleus.

Visual Map Representations in the Primate Pulvinar

The pulvinar receives direct visual information from the retina and indirect visual information from several cortical and subcortical areas. In this chapter, we discuss the visuotopic organization of the primate pulvinar. Electrophysiological techniques have been systematically employed to study pulvinar visuotopy in the owl, capuchin, and macaque monkeys. A single map of the visual field has been described in the pulvinar of the owl monkey, while two independent maps have been described in the capuchin and macaque pulvinar.

The Evolution of the Pulvinar Complex in Primates and Its Role in the Dorsal and Ventral Streams of Cortical Processing

Current evidence supports the view that the visual pulvinar of primates consists of at least five nuclei, with two large nuclei, lateral pulvinar ventrolateral (PLvl) and central lateral nucleus of the inferior pulvinar (PIcl), contributing mainly to the ventral stream of cortical processing for perception, and three smaller nuclei, posterior nucleus of the inferior pulvinar (PIp), medial nucleus of the inferior pulvinar (PIm), and central medial nucleus of the inferior pulvinar (PIcm), projecting to dorsal stream visual areas for visually directed actions. In primates, both cortical streams are highly dependent on visual information distributed from primary visual cortex (V1). This area is so vital to vision that patients with V1 lesions are considered "cortically blind". When the V1 inputs to dorsal stream area middle temporal visual area (MT) are absent, other dorsal stream areas receive visual information relayed from the superior colliculus via PIp and PIcm, thereby preserving some dorsal stream functions, a phenomenon called "blind sight". Non-primate mammals do not have a dorsal stream area MT with V1 inputs, but superior colliculus inputs to temporal cortex can be more significant and more visual functions are preserved when V1 input is disrupted. The current review will discuss how the different visual streams, especially the dorsal stream, have changed during primate evolution and we propose which features are retained from the common ancestor of primates and their close relatives.

An ascending visual pathway to the dorsal telencephalon through the optic tectum and nucleus prethalamicus in the yellowfin goby Acanthogobius flavimanus (Temminck & Schlegel, 1845)

Dual visual pathways reaching the telencephalon appear to be an ancient vertebrate trait, but some teleost fish seem to possess only one pathway via the optic tectum. We undertook the present study to determine if and when this loss occurred during evolution. Tracer injection experiments to the optic nerve, the optic tectum, and the dorsal telencephalon were performed in the present study, to investigate ascending visual pathways to the dorsal telencephalon in an acanthopterygian teleost, the yellowfin goby Acanthogobius flavimanus (Temminck & Schlegel, 1845). We confirmed the presence of a nucleus prethalamicus (PTh) in the goby, which has been convincingly identified only in holocentrids, suggesting that this nucleus is present in other acanthopterygians. We found that the optic tectum projects to the PTh bilaterally. The PTh projects in turn to the dorsal telencephalon, ipsilaterally. These results suggest that the yellowfin goby possesses only an extrageniculate-like pathway, while a geniculate-like pathway could not be identified. This situation is common with that of holocentrids and may be a character common in acanthopterygians. It is possible that a geniculate-like system was lost in the common ancestor of acanthopterygians, although the scenario for the evolution of ascending visual systems in actinopterygians remains uncertain due to the lack of precise knowledge in a number of actinopterygian taxons.

Unravelling the subcortical and retinal circuitry of the primate inferior pulvinar

The primate visual brain possesses a myriad of pathways, whereby visual information originating at the retina is transmitted to multiple subcortical areas in parallel, before being relayed onto the visual cortex. The dominant retinogeniculostriate pathway has been an area of extensive study, and Vivien Casagrande's work in examining the once overlooked koniocellular pathway of the lateral geniculate nucleus has generated interest in how alternate subcortical pathways can contribute to visual perception. Another subcortical visual relay center is the inferior pulvinar (PI), which has four subdivisions and numerous connections with other subcortical and cortical areas and is directly recipient of retinal afferents. The complexity of subcortical connections associated with the PI subdivisions has led to differing results from various groups. A particular challenge in determining the exact connectivity pattern has been in accurately targeting the subdivisions of the PI with neural tracers. Therefore, in the present study, we used a magnetic resonance imaging (MRI)-guided stereotaxic injection system to inject bidirectional tracers in the separate subdivisions of the PI, the superior layers of the superior colliculus, the retina, and the lateral geniculate nucleus. Our results have determined for the first time that the medial inferior pulvinar (PIm) is innervated by widefield retinal ganglion cells (RGCs), and this pathway is not a collateral branch of the geniculate and collicular projecting RGCs. Furthermore, our tracing data shows no evidence of collicular terminations in the PIm, which are confined to the centromedial and posterior PI.

Connectivity of the Pulvinar

Pulvinar connectivity has been studied using a variety of neuroanatomical tracing techniques in both New and Old World monkeys. Connectivity studies have revealed additional maps of the visual field other than those described using electrophysiological techniques, such as P3 in the capuchin monkey and P3/P4 in the macaque monkey. In this chapter, we argue that with increasing cortical size, the pulvinar developed new functional subdivisions in order to effectively interconnect and interact with the cortex.

Retinal afferents synapse with relay cells targeting the middle temporal area in the pulvinar and lateral geniculate nuclei

Considerable debate continues regarding thalamic inputs to the middle temporal area (MT) of the visual cortex that bypass the primary visual cortex (V1) and the role they might have in the residual visual capability following a lesion of V1. Two specific retinothalamic projections to area MT have been speculated to relay through the medial portion of the inferior pulvinar nucleus (PIm) and the koniocellular layers of the dorsal lateral geniculate nucleus (LGN). Although a number of studies have demonstrated retinal inputs to regions of the thalamus where relays to area MT have been observed, the relationship between the retinal terminals and area MT relay cells has not been established. Here we examined direct retino-recipient regions of the marmoset monkey (Callithrix jacchus) pulvinar nucleus and the LGN following binocular injections of anterograde tracer, as well as area MT relay cells in these nuclei by injection of retrograde tracer into area MT. Retinal afferents were shown to synapse with area MT relay cells as demonstrated by colocalization with the presynaptic vesicle membrane protein synaptophysin. We also established the presence of direct synapes of retinal afferents on area MT relay cells within the PIm, as well as the koniocellular K1 and K3 layers of the LGN, thereby corroborating the existence of two disynaptic pathways from the retina to area MT that bypass V1.

The retinal input to calbindin-D28k-defined subdivisions in macaque inferior pulvinar

Several studies have provided evidence for direct retinal input to the pulvinar of macaques monkeys, but there is no general agreement regarding the extent of this projection. Moreover, it is not known how retinal input correlates with chemoarchitectonic subdivisions recently recognized within the large, classical divisions of the pulvinar. The potential implications of this correlation have become more evident after reports that chemoarchitectonic subdivisions of the inferior pulvinar (PI) have specific patterns of connections with cortical visual areas. We have therefore re-examined the retino-PI projection using intraocular injections of horseradish peroxides, and correlated it with pulvinar subdivisions revealed using an antibody for calbindin-D28k. Retinal projections were found preferentially within the medial subdivision of the PI, with some involvement of the posterior and central calbindin-D28k defined subdivisions.

Viewpoint: the core and matrix of thalamic organization

The integration of the whole cerebral cortex and thalamus during forebrain activities that underlie different states of consciousness, requires pathways for the dispersion of thalamic activity across many cortical areas. Past theories have relied on the intralaminar nuclei as the sources of diffuse thalamocortical projections that could facilitate spread of activity across the cortex. A case is made for the presence of a matrix of superficially-projecting cells, not confined to the intralaminar nuclei but extending throughout the whole thalamus. These cells are distinguished by immunoreactivity for the calcium-binding protein, D28K calbindin, are found in all thalamic nuclei of primates and have increased numbers in some nuclei. They project to superficial layers of the cerebral cortex over relatively wide areas, unconstrained by architectonic boundaries. They generally receive subcortical inputs that lack the topographic order and physiological precision of the principal sensory pathways. Superimposed upon the matrix in certain nuclei only, is a core of cells distinguished by immunoreactivity for another calcium-binding protein, parvalbumin, These project in highly ordered fashion to middle layers of the cortex in an area-specific manner. They are innervated by subcortical inputs that are topographically precise and have readily identifiable physiological properties. The parvalbumin cells form the basis for sensory and other inputs that are to be used as a basis for perception. The calbindin cells, especially when recruited by corticothalamic connections, can form a basis for the engagement of multiple cortical areas and thalamic nuclei that is essential for the binding of multiple aspects of sensory experience into a single framework of consciousness.

Cell-specific expression of type II calcium/calmodulin-dependent protein kinase isoforms and glutamate receptors in normal and visually deprived lateral geniculate nucleus of monkeys

In situ hybridization histochemistry and immunocytochemistry were used to map distributions of cells expressing mRNAs encoding alpha, beta, gamma, and delta isoforms of type II calcium/calmodulin-dependent protein kinase (CaMKII), alpha-amino-3-hydroxy-5-methyl-4-isoxazoleproprionate (AMPA)/ kainate receptor subunits, (GluR1-7), and N-methyl-D-aspartate (NMDA) receptor subunits, NR1 and NR2A-D, or stained by subunit-specific immunocytochemistry in the dorsal lateral geniculate nuclei of macaque monkeys. Relationships of specific isoforms with particular glutamate receptor types may be important elements in neural plasticity. CaMKII-alpha is expressed only by neurons in the S laminae and interlaminar plexuses of the dorsal lateral geniculate nucleus, but may form part of a more widely distributed matrix of similar cells extending from the geniculate into adjacent nuclei. CaMKII-beta, -gamma, and -delta isoforms are expressed by all neurons in principal and S laminae and interlaminar plexuses. In principal laminae, they are down-regulated by monocular deprivation lasting 8-21 days. All glutamate receptor subunits are expressed by neurons in principal and S laminae and interlaminar plexuses. The AMPA/kainate subunits, GluR1, 2, 5, and 7, are expressed at low levels, although GluR1 immunostaining appears selectively to stain interneurons. GluR3 is expressed at weak, GluR 6 at moderate and GluR 4 at high levels. NMDA subunits, NR1 and NR2A, B, and D, are expressed at moderate to low levels. GluR4, GluR6 and NMDA subunits are down-regulated by visual deprivation. CaMKII-alpha expression is unique in comparison with other CaMKII isoforms which may, therefore, have more generalized roles in cell function. The results demonstrate that all of the isoforms are associated with NMDA receptors and with AMPA receptors enriched with GluR4 subunits, which implies high calcium permeability and rapid gating.

Retinal ganglion cells labelled from the pulvinar nucleus in macaque monkeys

In order to study the distribution and morphological classes of retinal ganglion cells that can be retrogradely labelled from the pulvinar nucleus, we made two iontophoretic injections of horseradish peroxidase into the pulvinar in each hemisphere of five macaque monkeys. The retrogradely labelled ganglion cells projecting to or through the pulvinar nucleus were examined in retinal whole-mounts. They comprise all three major ganglion cell classes. Primate gamma cells formed the great majority of classifiable cells and, like the primate alpha cells that were found in much smaller numbers, they were already known to send axons to the superior colliculus and to the pretectal complex. In contrast, the primate beta cells were hitherto thought to project solely to the dorsal lateral geniculate nucleus. This primate beta cell projection to an extrageniculate target could account in part for the substantial number of primate beta cells that escape transneuronal retrograde retinal degeneration following striate cortical ablation, and might contribute to the residual visual sensitivity that survives destruction of striate cortex and the degeneration of the lateral geniculate nucleus.

Cytochrome oxidase patches: a new cytoarchitectonic feature of monkey visual cortex

In normal macaque monkeys a histochemical stain for cytochrome oxidase activity revealed a striking pattern of regularly spaced patches in primary visual (striate, area 17, V1) cortex. The patches were most obvious in layers II and III, but also in layers I, IV b, V and VI; only in layers IV c and IV a were they absent. The patches were oval shaped, about 250 by 150 pm and aligned into rows spaced about 350 pm apart. Along each row a patch was located about every 550 pm; often patches in neighbouring rows were aligned, creating a square array. T heir density was about one patch per 0.2 mm 2 (550 by 350 pm) in opercular cortex. The patches were also labelled preferentially by stains for lactate dehydrogenase, succinate dehydrogenase, acetylcholinesterase (AChE), and myelin. In V2, a coarser pattern of broad parallel stripes labelled by cytochrome oxidase, lactate dehydrogenase, and AChE was present. The cytochrome oxidase patches were absent in non-prim ate species like the cat, mink, tree shrew, mouse, rat, rabbit, and ground squirrel. However, they were present in all prim ate species examined, including the rhesus, cynomolgus, owl, and squirrel monkey, baboon, bushbaby, and hum an. While more species should be tested, it appears that the patches are a cytoarchitectonic feature unique to prim ate visual cortex. In the owl monkey patches of anterogradely transported horseradish peroxidase (HRP) were found in layers IV c a , III, and II after injection of the tracer into the lateral geniculate nucleus (l.g.n.). They coincided exactly with the position of patches in adjacent sections processed for cytochrome oxidase. A similar result was obtained in the macaque, except that patches were not present in layer IV c a . These experiments established that the cytochrome oxidase patches receive a direct, patchy projection from the lateral geniculate body. However, retrogradely filled layer VI cells in the owl monkey bore no regular relation to the patches. In the macaque, the ‘honeycomb’ of geniculate terminals in layer IV a matched a similar honeycomb pattern of cytochrome oxidase staining. In the Nissl stain three sublayers in layer IV a were identified: the honeycomb was located in layer IV a p . In V2, in the owl monkey the parallel stripes of enhanced cytochrome oxidase activity received a direct projection from l.g.n. or pulvinar. In the macaque, after intraocular injection of [ 3 H]proline, the rows of patches in layers II and III lay in register with ocular dominance columns seen by transneuronal radioautography in layer IV c. In another macaque, one eye was removed and the cortex stained for cytochrome oxidase, AChE and Nissl substance after six months survival. In layer IV c light and dark bands corresponding to the ocular dominance columns were visible; surprisingly the dark cytochrome oxidase bands matched the light AChE and Nissl bands. The set of bands belonging to the missing eye was determined by examining cytochrome oxidase staining and proline radioautographs in another macaque that sustained severe eye injury by [ 3 H]proline injection. In striate cortex, bands of radioactive label from the injured eye matched ocular dominance columns appearing more lightly stained by cytochrome oxidase. In the macaque tested six months after enucleation, in every other row the cytochrome oxidase patches appeared pale and shrunken. These lighter rows fit into precise register with the lighter ocular dominance columns in layer IV c, confirming the correspondence between rows of patches and ocular dominance columns demonstrated by proline injection. AChE staining of patches was similarly affected by eye removal. The effect of visual deprivation upon cytochrome oxidase staining was tested in two monocularly sutured macaques. In the l.g.n. no effect was detected. In visual cortex wide light columns alternating with thin dark columns were observed in layer IV. In one m acaque the ocular dominance columns were labelled independently by H R P injection into a deprived l.g.n. lamina. The H R P labelled ocular dominance columns fit within the pale cytochrome oxidase columns; this establishes that monocular deprivation causes a relatively greater loss of enzyme activity in ocular dom inance columns belonging to the closed eye. However, there was also loss of cytochrome oxidase staining along the borders of the normal eye dominance columns, indicating that ocular dominance columns in layer IV are subdivided into core zones flanked by border strips that are susceptible to loss of cytochrome oxidase activity with suture of either eye. The core zones are the same width as the rows of cytochrome oxidase patches and correspond to the dark bands seen in Liesegang stains of normal macaque striate cortex. In two adult cats the effect of monocular lid suture at 28 d old was assessed: no effect upon cytochrome oxidase staining in l.g.n. or cortex was observed. The optic disc representation in visual cortex was studied by 2-deoxyglucose radioautography and cytochrome oxidase staining after eye removal or lid suture in m acaque monkeys. It appeared as a pale oval, 1.65 times longer than the optic disc, a distortion probably required to m aintain overall isotrophy in magnification factor. Patches were present in the disc representation although ocular dominance columns are absent: they appeared rounder and more widely separated. In the temporal cresent patches were also present. They were larger, rounder, and less densely spaced than patches in binocular cortex. Deoxyglucose mapping in a macaque monkey monocularly stimulated with a display of parallel black and white stripes of irregular width and spacing rotated through all orientations has resulted in patches in the upper layers over ocular dominance columns corresponding to the open eye. These patches match cytochrome oxidase patches situated in every other row, thus suggesting that cells located in cytochrome oxidase patches respond to all orientations of stimulus. Macaques binocularly stimulated with vertical or horizontal stripes show a complicated pattern of deoxyglucose uptake, overlapping extensively with the pattern of cytochrome oxidase patches. In one monkey the right eye was removed and 18 d later the animal was stimulated with vertical stripes. Deoxyglucose radioautography and cytochrome oxidase staining combined in single tissue sections each revealed a matching pattern of ocular dominance columns in layer IV. In the upper layers, dots of radioautographic label were present, matching cytochrome oxidase patches in alternate rows. In foetal monkeys at E142-144 the laminar pattern of cytochrome oxidase staining in visual cortex was remarkable for a prominent wide band of intense activity in layer IV b and upper IV c a , absent in mature macaques. In tangential section, patches were visible in layers II, III and in layer IV b -IV c a , which indicates that patches form in monkey visual cortex before birth. The functional significance of the patches remains uncertain. It has been suggested that the visual field is analysed in visual cortex by small modules containing several hypercolumns of each stimulus variable. The cytochrome oxidase patches may constitute the anatomical correlate of these proposed modules.

Retinal projections to the pulvinar nucleus of the macaque monkey: a re-investigation using autoradiography

After a monocular injection of [3H]amino acid into the vitreous chamber of the eye, the distribution of retinal terminations in the pulvinar nucleus of the crab-eating monkey and pigtail macaque was studied by autoradiography. Two groups of labeled terminals were found in the bilateral inferior pulvinar nuclei: one small, dense group was located in the most rostral part of the nuclei and the other, composed of a few small clusters of the labeled terminals, was observed over the medial zone of the middle portion. The terminals were slightly predominant in the contralateral nucleus. A small amount of silver grains showing labeled retinofugal fibers was found in the dorsal surface of the thalamus just medial to the stria terminalis contralateral to the injected site, but termination of these fibers could not be traced in this study.

Retinal projections to the inferior and medial pulvinar nuclei in the Old-World monkey

Horseradish peroxidase (HRP) and wheat germ agglutinin conjugated to HRP (WGA-HRP) were used as anterograde tracers to study the macaque monkey retinofugal pathways. Labeled axons were mapped beyond the lateral geniculate nucleus to the inferior (PI) and medial (PM) pulvinar nuclei, where they terminated. Retinofugal fibers project to the medial part of PI, adjacent to the brachium of the superior colliculus. A smaller projection to PM leaves the ventrolateral aspect of the lateral geniculate nucleus and travels parallel to the stria terminalis to reach the dorsal surface of the thalamus and PM. HRP reaction product was bilateral in PI and predominantly contralateral in P)M.

A retino-pulvinar projection in the macaque monkey as visualized by the use of anterograde transport of horseradish peroxidase

A direct retino-pulvinar connection was found in the Japanese monkey (Macaca fuscata) by the anterograde horseradish peroxidase (HRP) method. HRP injected into the vitreous cavity of one eye labeled a few small clusters of optic fiber terminals in the medial border regions of the inferior pulvinar nucleus bilaterally, with a contralateral predominance.

An experimental electron microscopical study of a direct retino-pulvinar pathway in the tree shrew

After unilateral enucleation in the tree shrew (Tupaia glis), axo-dendritic synapses degenerate within two secondary visual centers (intergeniculate and lateral pulvinar nuclei). Degenerated optic terminals were found in both pulvinar nuclei on the contralateral side much more often than on the ipsilateral one.

A retino-pulvinar projection in the cat

Injections of tritiated amino acids were made in one eye of Siamese and common cats including young kittens. After survival periods of 1--7 days axoplasmically transported label accumulated in a portion of the pulvinar nucleus as well as in the other known sites of termination of the retinofugal pathway. The retino-pulvinar projection is present at birth; it is bilateral and approximately symmetrical in common cats but the ipsilateral component is markedly reduced in Siamese animals. Labeled terminal ramifications of the retinal fibers in the pulvinar take the form of a thin, interrupted sheet oriented dorsoventrally and lying at the extreme lateral edge of the pulvinar nucleus. It appears to be continuous caudally with the medial interlaminar nucleus of the lateral geniculate complex, but the cells about which the grains cluster are clearly different from those of the medial interlaminar nucleus.

Retinofugal pathways in two marsupials

Autoradiographic and anterograde degeneration tracing methods were used to study and compare the organization of retinofugal pathways in two marsupial opossums, Didelphis virginiana and Marmosa mitis. Seven identical retinal targets were demonstrated for each opossum. These include: (1) the suprachiasmatic nucleus of the hypothalamus, (2) the dorsal and (3) ventral lateral geniculate nuclei, (4) the lateral posterior nucleus, (5) the pretectal complex, (6) the superior colliculus and (7) the accessory optic nuclei. While the pattern of retinal input to six of the seven targets was quite similar in the two species, the organization of the retinogeniculate pathways exhibited striking differences. In particular, our autoradiographs reveal no separation of ocular inputs within the lateral geniculate nucleus of Didelphis, i.e. the ipsilateral input is overlapped completely by the more extensive contralateral projection. In contrast, there is considerable separation, as well as overlap, of the occular inputs within the lateral geniculate nucleus of Marmosa. Our autoradiographs reveal several distinct bands of label within each geniculate nucleus, and upon superimposing the nuclei, ipsilateral and contralateral to the placement it is apparent that two of the bands overlap, while five do not (three ipsi, two contra).

Behavioural analysis of the monkey's visual nervous system

1. Introduction Of the three million or so nerve fibres that stream into the primate brain, about two million originate in the eyes. Of these fibres, about one-and-a-half million are in the geniculo-striate system, so named because it connects the eyes with a region of the thalamus known as the dorsal lateral geniculate nucleus and that nucleus with the striate cortex (also known as area 17 or area OC, figure 1) in the occipital lobes. About half, therefore, of all the inputs to the brain are fibres of retinal origin having relatively direct and concentrated access to the cerebral cortex. One may be allowed some surprise, therefore, to find that David Ferrier claimed in 1886 that monkeys subjected to large occipital lobectomies (figures 2, 3) were unaffected by this drastic interruption of such a massive afferent channel. He said 'I removed the greater portion of both occipital lobes at the same time without causing the slightest appreciable impairment of vision. One of these animals within 2 h of the operation was able to run about freely, avoiding obstacles, to pick up such a minute object as a raisin without the slightest hesitation or want of precision, and to act in accordance with its visual experience in a perfectly normal manner’ (Ferrier 1886, p. 273). Ferrier went on to say that ‘Horseley and Schäfer inform me that their results of removal of the occipital lobes entirely harmonize with mine as to the completely negative effect of this operation’ (p. 276), which is a curious claim because two years later Schäfer was locked in a most bitter dispute with Ferrier over just this point, and their argument is merely the most extreme example of the lack of agreement about the functions of the visual cortex in animals that has persisted over the years. We now know, with the benefit of hindsight, that there may have been an uninteresting explanation of these early results of Ferrier’s, because not all of the fibres from the lateral geniculate nucleus project to the lateral surface of the brain. Some of the striate cortex ─ that part which responds to stimulation of the most peripheral parts of the retinae ─ is buried in the calcarine fissure on the medial aspect of the brain, and the most anterior portion of this may be spared even after a complete occipital lobectomy (figure 1). Were Ferrier’s animals using an intact part of their visual space?