sign in sign up
<<
Home , Fundus (eye), Ganglion cell layer, Retina

The Naso-Temporal Division of the Monkey’s

JONATHAN STONE. JONATHAN LEICESTER AND S. MURRAY SHERMAN* Department of Physiology, John Curtin School of Medical Research, Australian National Untversity, Canberra

ABSTRACT By sectioning one in each of four monkeys, and studying the distribution within each pair of of the ganglion cells which remained after the affected ganglion cells had undergone retrograde degenera- tion, a description was obtained of the areas of a retina from which ganglion cells project to the ipsilateral and contralateral sides of the . Confirming previous work, ganglion cells in temporal retina were observed to project to the ipsilateral side. and cells in nasal retina to the contralateral side. It was noted, however, that ipsi- and contralaterally projecting areas of retina overlap along a vertically-oriented median strip, which is about 1 wide, and is centred on the fovea. Within this strip ipsi- and contralaterally projecting cells intermingle. Some functional implications of this result are discussed.

Each retina of the monkey projects to removed by suction (fig. 1) allowing visu- both ipsi- and contralateral sides of the alization of the optic , and brain (Polyak, ’57; Kupfer, ’63; van Buren, tract. A hook was passed under the tract ’63). Specifically, the ganglion cells located and pulled upwards, dividing it. Surgery in the area of retina nasal to the fovea send was performed in adult or young adult their to the contralateral optic tract, (2-year-old) monkeys. and cells located temporal to the fovea Following survival times of 6 to 12 send their axons to the ipsilateral optic months the monkeys were anaesthetized tract. This study is an attempt to define and perfused through the heart with 0.9% the border between ipsi- and contralater- saline followed by 10% formol-saline. Two ally projecting areas of retina. Evidence is animals were, before perfusion, placed in presented that the borders of these two a stereotaxic apparatus, paralyzed by in- areas run vertically across the fovea and travenous injection of gallamine triethio- that the two areas overlap to a small but dide (Flaxedil) and artificially respired. significant degree. As in the (Stone, Their fundi were photographed using a ’661, a median strip of ouerlap can be de- Zeiss , and the scribed within which ganglion cells which and fovea of each retina were projected project to ipsi- and contralateral optic onto a frontal 1 metre tangent screen, tracts intermingle. The strip runs vertically using techniques described previously in the retina, passing across the fovea. In (Bishop, Henry and Smith, ’71). Following the monkeys used (Macacca irus) its width perfusion the of each animal were ranged from 150-250 pm. This corre- enucleated and the brain removed. The sponds to approximately 1 O of visual angle. eyes of the two animals whose fundi had been photographed were placed in saline METHODS and photographed with eyeball gently in- The experimental approach was basic- flated. These photographs allowed estima- ally that used for the cat (Stone, ’66 >. Four tion of the radii of curvature of the monkeys (Macncca irus) were used. In and eyeball and of the length of the - each one optic tract was sectioned by ball. The opened fundus of each of these aseptic surgery under pentobarbitone an- 1 Present address: Suite 20A, “Murray Place,” 127 aesthesia. The approach was through the Burwood Road, Burwood, 2134 N.S.W. temporal bone, with the animal lying on 2 Present address: Department of Physiology, School of Medicine, University of Virginia, Charlottesville, its side. The tip of the temporal lobe was Virginia.

J. COMP NEUR,150 333-348. 333 3 34 J. STONE, J. LEICESTER AND S. M. SHERMAN

Fig. 1 Ventral surface of brain of tract-sectioned monkey. The dotted line marks the surgical approach, through the right temporal lobe. The arrow points to the peripheral stump of the right optic tract. four eyes was subsequently photographed the peripheral pieces were also mounted (fig. 10). and stained. The staining technique was a Methylene blue-stained whole mounts of modification of that described previously the retinas were prepared as follows. Each (Stone, '65, '66) for whole mounts. eye was cut around so that the anterior It is somewhat simpler than the earlier part (, , and cornea) techniques, and has proved entirely re- was separated from the posterior, fundal liable. The preparations have now lasted part. The anterior part was discarded and two years without deterioration, and ap- the posterior half placed in 10% formol- pear likely to last indefinitely. Each piece saline for two days. The posterior half of of retina was dissected free of the . each eye was then cut into appropriate In the central piece the was pieces. Particular attention was given to cut with the tip of a scalpel blade passed the central piece, which was always cut between retina and choroid. Each piece of to include the optic disc and the fovea, but retina was cleaned as completely as pos- NASO-TEMPORAL DIVISION OF MONKEY RETINA 335 sible of pieces of choroid clinging to its by placing for three minutes in each of outer surface and of the vitreous humour the following solutions : 90% alcohol, attached to its inner surface. The piece was absolute alcohol (2 changes) and xylol then laid, fibre layer uppermost, on a (2 changes). gelatinized glass slide and processed 5. Depex was used for mounting. through the follow.ing steps : This sequence is equally effective for 1. Each preparation (i.e., each piece of staining cat and rabbit retina; we com- retina spread on a gelatinized glass slide) monly processed a piece of cat retina im- was placed in hot (60°C) formalin vapour mediately before the monkey retinas, to for two hours, so that the retina stuck check for defective solutions. firmly on the slide. The remaining steps were done at room temperature. RESULTS 2. The preparation was washed briefly Optic tract sections in distilled water, and then stained by lay- ing the retina “face-down’’ over a small Figure 1 shows the appearance of the Petrie dish brimming with 0.2% methylene ventral surface o€ the brain of one tract- blue. This face-down contact of retina with sectioned subject. The arrow follows the stain minimized the precipitation of par- surgical approach through the temporal ticles of stain onto the retinal surface. lobe and points to the stump of the Three minutes staining time at room tem- severed optic tract. Essentially identical perature reliably gave an appropriate depth controls of the tract section were obtained of stain, but this can be monitored under in the other animals. a microscope. 3. Excess stain was wiped off and the Normal retina stain was fixed in 5% ammonium molyb- In the normal monkey retina, seen in a date (3 minutes). whole mount preparation (fig. 2), the 4. The preparation was washed in dis- is a circular area about 500 pm in tilled water, then dehydrated and cleared diameter, which is surrounded by densely

Fig. 2 Foveal area of the right retina of a normal Macncca irus seen in a methylene blue- stained whole mount. The pale circle in the centre of the field is the foveola, i.e. the area free of ganglion and bipolar cells. The ganglion cells form a continuous layer, several deep, surrounding the foveola. The paths of bundles and of small blood vessels are apparent as faint streaks and loops. The arrow points to the optic disc. 336 J. STONE, J. LEICESTER AND S. M. SHERMAN packed ganglion cells (see also fig. 10 in Figure 5 shows, at still higher power, Stone, ’65). Pale streaks are apparent areas of the at the running across the surfacc of the ganglion junction of normal and degenerated areas. cell layer and revealing the pattern of Figure 5A shows an area at the upper fibre bundles around the foveola (cf. fig. margin of the foveola (labelled F) of the 162 in Poliak, ’57). The streaks mark the right retina shown in figures 3A and 4A. paths of fibre bundles. They appear be- Figure 5B shows an area at the lower cause the fibre bundles limit access of the margin of the foveola of the left retina stain to the underlying ganglion cells. The shown in figures 3B and 4B. In each of more tortuous pale lines are the paths of figures 5A and 5B a sharp gradient in superficial blood vessels. ganglion cell density is apparent. Many cell bodies are nevertheless apparent in areas The retinal distributions of ipsi- and lacking ganglion cells; their significance contralaterally projecting ganglion is considered further below. cells: basic observations Figures 6 and 7 show the whole mounts Confirming previous observations in obtained from two other animals. In each monkey and man (Gartner, ’51; Kupfer, eye the line dividing normal and degener- ’63; van Buren, ’63) ganglion cells in the ated retinal areas could be traced for 2 to temporal half of the retina ipsilateral to the 3 mm above and below the fovea. Because tract section and in the nasal half of the the line is formed by ganglion cells it be- contralateral retina underwent retrograde comes less distinct towards the retinal degeneration. Figure 3A shows at low periphery, as the density of ganglion cells power the appearance of a methylene blue decreases. stained whole mount of the right (ipsilat- The overlap eral) pztina of one monkey, six months In micrographs such as figures 3, 4, following section of the right optic tract. 6 and 7 it is reasonably straightforward to Up on the figure is “up” on the retina and delineate the areas of retina which contain left on the figure is “temporal” on the ipsi- and contralaterally projecting gan- retina; i.e. the retina is “looking at” the glion cells. Do these areas overlap in one reader. This convention is followed in all retina? Figure 8A was traced from the relevant drawings and illustrations. The right retina in figure 4A. The foveola is fovea is in the centre of the illustration outlined, the area containing ganglion and the surrounding retina can clearly be cclls is hatched and its temporal limit is divided into the lighter, temporal (left) drawn as a line. Figure 8B is the analogous half, in which the ganglion cells have de- tracing obtained from the left retina in generated and disappeared and the darker, figure 4B. Figure 8C is figure 8B reversed nasal (right) half in which ganglion cells left-to-right. Figure 8A then represents appear to be present in normal numbers. the area of a right retina containing con- Figure 3B shows the appearance of the tralaterally projecting ganglion cells, and left retina of the same animal. It is figure 8C represents the area of a right equally clearly divided into areas lacking retina containing ipsiluterully projecting and containing ganglion cells. Figures 4A cells. When figures 8A and 8C are super- and B show the foveal areas of these imposed, with the outlines of the foveolae retinas at higher magnification. In each superimposed as closely as possible, figure retina the boundary of the area conlain- 8D is obtained. The areas of ipsi- and ing ganglion cells is an approximately contralaterally projecting cells appear to straight line which passes slightly to one overlap along a narrow (approximately side of the midpoint of the foveola. The 150 pm wide) vertical strip which runs line can be considered to run vertically be- across the centre of the fovea. Figures 9A, cause it runs perpendicular to the line B, C show the strip of overlap of ipsi- and joining the fovea to the disc (see figs. 3, contralaterally projecting areas of retina lo). This latter line runs horizontally, as obtained in three other monkeys. The strip assessed ophthalmoscopically, and by pro- is clearly analogous to the median strip jection of retinal landmarks in the of overlap described in the cat’s retina paralyzed animal [table 1). (Stonc, ’66). NASO-TEMPORAL DIVISION OF MONKEY RETINA 337

Fig. 3 Retinas of onc tract-sectioned monkey seen in methylene blue whole mounts. A: Right retina: The fovea is at the centre of the photograph and the optic disc is the torn patch at right, from which blood vcssels and axon bundles radiate. The area of retina temporal to (to the left of) the fovea is lacking ganglion cells and has stained relatively lightly. The area of retina nasal to (to the right of) the fovea has a normal population of ganglion cells, and has stained more darkly. The border between the two areas is a straight line which runs vertically in the retina, crossing the fovea. The border runs perpendicular to the line joining fovea to the centre of the optic disc. €3: Left retina: The fovea is at the centre of the figure and the optic disc is at left. Ganglion cells are lacking from the area of retina nasal to the fovea and are present in the area temporal to the fovea. The border between these two areas is a straight line which runs vertically in the retina, cross- ing the fovea. In both left and right retinas the border between areas containing the lacking ganglion cells runs slightly to the left of the centre of the foveosla. For further interpretation, see text. 338 J. STONE, J. LEICESTER AND S. M. SHERMAN

Figs. 4A.B Foveal regions of the retinas of figure 3A,B respectively, at higher magnifica- tion. In each retina the arrow points to the optic disc.

Ophthalmoscopic and post-mortem is due to the absence of axon bundles appearance of the fundus from such areas. In the right retina (fig. Ophthalmoscopically the only suggestion 10A) the fibre bundles form a continuous of abnormality was a sheen on the areas of layer of axons as they converge on the retina from which ganglion cells had de- optic disc, as in the normal retina, hut the generated. This was more apparent in the bundles begin to form only nasal to the right retina, in which the temporal half of fovea. A clear line between normal (nasal) the retina was affected. The area of sheen and degenerated (temporal) areas of appeared to include the lateral margin of retina is apparent for 2-3 mm above and the fovea. below the fovea. In the left retina (fig. When the eye was opened post-mortem lOB), the axon bundles all arise from clear indication of the pattern of ganglion ganglion cells in temporal retina and form cell degeneration was apparent, even be- two arcuate streams which course above fore staining (fig. lo). The retina appeared and below the disc-fovea line. A clear line “milky” and opaque where it was normal, between normal (temporal) and degener- and was more transparent, making the ated (nasal) areas of retina is apparent fundus appear darker, where the ganglion for only about 1 mm above and below the cells had degenerated. The impression was fovea. Further up and down axon bundles gained that the relative transparency of cross the line, obscuring it. The same effect the retinal areas affected by degeneration is seen in the stained retina (fig. 3B), be- NASO-TEMPORAL DIVISION OF MONKEY RETINA 339

TABLE 1 Parameters of monkey eyes

Monkey 1 Monkey 2 Left eye Right eye Left eye Right eye

Optic disc : Vertical axis 1.6 mm, 8.4" 1.6 mm, 8.0" 1.45 mm, 7.5" 1.45 mm, 8.0' Horiz. axis 1.0 mm, 5.2" 1.15 mm. 5.6" 1.0 mm, 5.2" 1.10 mm, 5.9" Disc centre to fovea 3.2 mm, 18" 3.2 mm, 16" 3.2 mm, 18" 3.25 mm, 18" PND 11.1 mm 11.5 mm 11.0 mm 10.5 mm (range) (10.4-11.9 mm) (11.3-11.6 mm) (10.5-11.4 mm) (10.2-10.7 mm)

Radius of corneal 5.2mm 5.2 mm 5.5 mm 5.5mm curvature Antero-post. length of 17.5 mm 17.5 mm 17.0 mm 17.0 mm eyeball Angle between the horizontal and the line 3" 1" 4" 3" joining blind- spot to fovea, in paralyzed animal cause the axon bundles, although they do Observations on the cells remaining not stain, reduce the access of the stain in the ganglion cell layer after to the underlying ganglion cells. tract section Are there contralaterally projecting It was important to test whether the cells which remained in the ganglion cell layer, cells in temporal retina? in areas of retina such as those illustrated Stone ('66) concluded that in the cat in figure 11 included ganglion cells whose 25% of the ganglion cells of temporal axons somehow escaped section. Many of retina project to the contralateral optic these remaining cells are much smaller tract, and the presence of contralaterally than ganglion cells but some are of com- projecting cells in temporal retina in the parable size. Figure 11 shows a compari- cat has been confirmed physiologically son of degenerating areas of nasal retina (Ogawa et al., '69; Fukada, '71). Such (upper row) and temporal retina (lower cells do not appear to be present in the row). The areas in figures 1la and d are on monkey retina. Figures 1la,d show respec- the margin of the foveola; the areas in b tively the nasal (degenerated) margin of and e are 2.0 mm from the fovea; the areas the left foveola and the temporal (degener- in c and f are 5.0 mm from the fovea. ated) margin of the right foveola, from one The density of cells does not decrease con- monkey. Numerous cell bodies remain in sistently with the increasing distance from both areas but evidence is presented below the fovea, as might be expected if a con- that few, if any, are ganglion cell bodies. siderable proportion of the cells were gan- Importantly, the impression is not gained glion cells. Interestingly, the highest that, as in the cat, there are more and density of cells is in the area in figure 1 le. larger cell bodies remaining in the tem- This area is 2.0 mm nasal to the fovea, and poral area. The same similarity is appar- hence about 0.5 mm from the edge of the ent (figs. llb,c,e,f) between peripheral optic disc. The higher density of cell bodies areas of nasal and temporal retina (both in this area can be explained by the as- degenerated). sumption that these are the bodies of 340 J. STONE, J. LEICESTER AND S. M. SHERMAN

Fig. 5 A: Area at the upper margin of the fovea of the retina in figures 3A and 4A, at higher magnification. The plane of focus is at the gangiion cell layer, showing the gradient in ganglion cell density in this region. B: Area at the lower margin of the fovea of the retina in figures 3B and 4B. The plane of fcous is at the ganglion cell layer. The lctters F mark the position of the edge of the foveola, in each of A and B. neuroglial cells, which increase in num- Appearance of the retina in sections ber near the optic disc to provide support for the axon bundles of the fibre layer, One retina (that shown in whole mount which reaches its maximal thickness near in fig. 7A) was carefully detached from the disc. A second test of the nature of the the slide with a razor blade, embedded in cells which remain in the ganglion cell celloidin, sectioned and restained with layer after tract section would be provided cresyl violet (the methylene blue was by comparison with a retina from which washed out by various alcohols). On the all ganglion cells had degenerated, follow- unaffected side of the retina the ganglion ing optic nerve section. Towards the end cell, inner nuclear and outer nuclear layers of the present study, optic nerve section are clearly visible both at the fovea (fig. was attempted in two animals. One animal 12B) and 5 mm away (fig. 12D). On the survived surgery but, when examined six side affected by tract section the number months later, the section proved incom- of cell bodies in the ganglion cell layer is plete. A further attempt at complete nerve dramatically reduced (fig. 12A, C). The section has been made and will be re- inner and outer nuclear layers appear ported elsewhere. normal. NASO-TEMPORAL DIVISION OF MONKEY RETINA 341

Figs. 6A,B Right and left retinas respectively of a second monkey, presented in the same format and at the same magnification as figure 4.

Physical dimensions of the To a close approximation the posterior monkeys’ eyes nodal distance (PND) of the eye is given Two animals were paralyzed before per- by (Vakkur, Bishop and Kozak, ’63) fusion and fixed in a stereotaxic head- PND = T x d/D holder facing a tangent screen 1 m from the eye. The boundaries of the optic disc where T is the distance from the anterior of each eye were then projected onto the nodal point of the eye to the tangent tangent screen, giving an outline of the screen (1 m); . The centre of the fovea was D is a dimension on the tangent screen similarly projectcd giving the position of (say the distance from the projection of the point. In these two animals the fovea to the centre of the blind spot) several dimensions of the eyes were mea- and sured, from photographs of the eyes taken d is the corresponding distance on the immediately post-mortem. These dimen- retina (i,e., the distance from the fovea to sions were the antero-posterior length of the centre of the optic disc). the eyeball, the radius of corneal curva- For each eye three estimates of the PND ture, the size of the optic disc and the dis- were made, one using the disc-fovea dis- tance between the centre of the disc and tance, and the other two using the vertical the fovea. These measurements are tabu- and horizontal axes of the disc. The mean lated in table 1. and range of values for each eye are given 342 J. STONE, J. LEICESTER AND S. M. SHERMAN

Figs. 7A,B Right and left retinas respectively of a third monkey, presented in the same format and at the same magnification as figures 4 and 6. in table 1. For all four eyes the mean PND estimated that the shrinkage of the retina was 11.0 mm and the range 10.2 mm to during processing of the whole mount 11.9 mm. preparations was 5% of length or less. From the posterior nodal distance the This estimate was confirmed in the prep- angle subtended by the strip of overlap can arations made in this study. For example, be calculated. As already noted the strip, the distance between the centre of the as estimated here, varies in width from optic disc and the fovea was measured in animal to animal, and with position in two animals (4 eyes) with the retina in any given animal. In the schematic dia- situ, after the cornea and lens had been grams in figures 8D and 9 the strip varies removed (fig. 10). This distance ranged in width approximately from 0.5" to 1.25". from 3.20 to 3.25 mm (table 1). In the Considering the errors inherent in this whole mount preparations of these retinas, measurement (see DISCUSSION)it seems a this distance ranged from 3.1 mm to 3.2 reasonable approximation to state the mm (e.g., fig. 3). strip of overlap is about 1" wide. DISCUSSION Shrinkage of whole mount Sources of error preparations Two sources of error in the estimation In a previous study (Stone, '66) it was of the width of the strip of overlap must NASO-TEMPORAL DIVISI(IN OF MONKEY RETINA 343

centered on the fovea. It is approximately A: RIGHT RETINA 1 wide. Degenerated rmd Degree of fzrnctional overlap The present study describes only the Fovea f retinal distributions of ipsi- and contra- laterally projecting ganghon cells. Thus the overlap described is an overlap of the areas of retina which contain contra- and D: A and C superimposed: C. Left retina ipsilaterally projecting ganglion cells. What ' inverted left to right is the functional overlap? That is, how wide is the overlap of ipsi- and contra- laterally projecting receptors? Morphologi- cal studies (see for example, plate 34 in Boycott and Dowling, '69) indicate that, except at the fovea, the processes of re- ceptor and bipolar cells run radially through the thickness of the retina, con- necting the ganglion cells to the receptors immediately external to them. It seems Fig. 8 A: Schematic drawing of the foveal re- likely, therefore, that the overlap of ipsi- gion of the right retina in figure 3A. The outline and contralaterally projecting cells is of the foveola is drawn in and the areas lack- at ing and containing ganglion cells are open and least as wide in the receptor layer as in the hatched, respectively. The border between these ganglion cell layer. The overlap may be two areas is marked as a line. B: Analogous draw- somewhat wider in the receptor layer, since ing for the left retina of the same animal. C: The drawing in B reversed left-to-right. D: A and C many ganglion cells may have receptive superimposed. The outlines of the foveolae are fields larger than their cell bodies. matched as closely as possible and the lines are made as close to parallel as possible. This figure The problem of overlap at the fovea gives an estimate of the strip of overlap, marked here with double hatching. The presence of the fovea in monkey retina made it relatively simple to demon- strate the presence of a strip of overlap. be stated. First, the retina in vivo approxi- mates in shape the surface of a sphere. The task was certainly simpler than in the Consequently, the retina is necessarily dis- cat retina, where it was necessary to quan- torted somewhat as it is flattened onto a tify ganglion cell density. (Conversely the glass slide. This distortion was minimized high density and rnultilayering of gan- by keeping the piece of retina containing glion cells up to 8 rnm from the monkey fovea (van Buren, '63)makes the counting the optic disc and fovea fairly small (typic- approach used in the cat impractical in the ally 9 mm X 9 mm), but could not be monkey). However, because the ganglion eliminated. This has presumably caused cells connected to receptors within the some distortion in the shape of the strips foveola are displaced from those receptors, of overlap shown in figures 8 and 9. Sec- the present study does not test whether ond, there is presumably some small error there is overlap within the foveola. It does in figures 8 and 9 in the placement of the demonstrate that the strip of overlap in borders of the areas containing ganglion the ganglion cell layer runs, apparently cells. However, although these two factors without distortion, across the superior and clearly limit the accuracy with which the inferior margins of the fovea, which con- width of the strip of overlap can be esti- tain ganglion cells displaced upwards and mated, neither seems sufficient to qualify downwards, respectively, from the foveola. the conclusion that there is a strip of over- It does also demonstrate that contralater- lap, within which ipsi- and contralaterally ally-projecting ganglion cells are displaced projecting ganglion cells intermingle. The to the nasal margin of the foveola, and strip runs vertically in the retina, and is ipsilaterally-projecting cells to the tem- 344 J. STONE, J. LEICESTER AND S. M. SHERMAN

A. B. C.

lrnm lmrn- lmm lo Fig. 9 Drawings like that in figure 8D for three other monkeys. In each the strip of overlap is represented by the double-hatching.

Fig. 10 The appearance of the fundi of the right (A) and left (B) eyes of one monkev, immediately after perfusion. The stained retinae of these two eyes are shown in figures 3, 4 and 5. In each fundus the foveola is the dark spot at the centre of the figure. The optic disc is the white ellipse from which blood vessels radiate. poral margin. But because it is an analysis Functional significance of the positions of ganglion cell bodies the Linksz ('52) argued that ipsi- and con- present approach cannot demonstrate tralaterally projecting areas of human overlap within the foveola. retina must overlap, since a precise, non- It is, of course, conceptually economical overlapping wiring would be biologically to assume that there is no functional dis- unlikely. Ogle ('62) argued for the pres- continuity at the foveola in the overlap of ence of an overlap, on the grounds that ipsi- and contralaterally projecting areas Panum's area shows no discontinuity at the of retina, and this assumption is made in fixation point. Both workers recognized the following discussion. It may however that the overlap might be too small to be be important to keep in mind that it is an detectable by perimetry. The present study assumption. seems to confirm their ideas. NASO-TEMPORAL DIVISION OF MONKEY RETINA 345

Fig. 11 Upper row (a,b,c): Areas from the right retina of a monkey in which the right optic tract was sectioned (same retina as in figs. 3A, 4A, 5A). Figure lla shows an area on the margin of the foveola 0.4 mm temporal to the centre of thc foveola; b shows an area 2 mm temporal to the centre of the foveola; c shows an area 5 mm temporal. Lower row (d,e,f): Areas from the left retina of the same animal (same retina as in figs. 3B, 4B, 5B). Figure lld shows an area on the nasal margin of the foveola; e shows an area 2 mm nasal; f shows an area 5 mm nasal.

The presence of a central, vertically- monkey suggests that these theories may running strip of retina which projects to be equally valid for the monkey, and per- both hemispheres has become important haps for man. in recent theories of the neurophysiological Finally, the possibility must be consid- basis of binocular depth discrimination ered that the overlap of ipsi- and contra- (Blakemore, '69; Bishop, '70). These laterally projecting areas of retina might theories were developed from studies of form a basis for macular sparing. Although disparity in cat visual cor- the present results suggest that more than tex. The present finding that a strip of half the fovea is functional after tract sec- overlap of similar width is present in the tion, considerable difficulties arise in any 346 J. STONE, J. LEICESTER AND S. M. SHERMAN

Fig. 12: Sections of the retina shown in figure 7A. A: Temporal margin of the foveola; B: Nasal margin of the foveola; C: Retina 5 mm temporal to the fovea; D: Retina 5 mm nasal to the fovea; g.c.1.: ganglion cell layer; i.n.1.: ; o.n.1.: . attempt to account for macular sparing in of the hemifield also retained function. terms of the nasotemporal overlap of They also noted that macular sparing could retina. For instance, Williams and Gassel be induced, or abolished, by varying the (’62) describe two broad types of macular technique of perimetry. They concluded sparing, “true” and “false.” In true macu- that the occurrence of true macular spar- lar sparing, vision appears to be main- ing depends on the extent of the lesion, the tained in the macular area of a hemifield attention of the subject and the technique rendered otherwise blind by a brain lesion, of perimetry applied. False macular spar- usually in or including . They ing was the term applied by Williams and noted, however, that the spared central Gassell to apparent survival of visual func- area may be 3-4” wide, much wider than tion in a 3-4” wide strip of an otherwise the strip of overlap, and that in all the blind hemifield located adjacent to the patients they examined some other part vertical meridian of the . These NASO-TEMPORAL DIVISION OF MONKEY RETINA 347

workers report that survival of vision depth discrimination. J. Physiol. Lond., 163: within the strip was always clearly attribut- 327-342. Bishop, P. 0. 1970 Begicning of foim vision able to eccentric fixation. Thus macular and binocular depth discrimination in cortex. sparing as currently understood, does not In: The : Second Study Program. seem to be clearly related to the approxi- F. P. Schmiti, ed. Rockefeller Univ. Press, New mately 1O wide bilaterally-projecting strip York, N.Y. of retina described in this study. Bishop, P. O., G. H. Henry and C. J. Smith 1971 Binocular interation fields of single The central projection of units in cat striate cortex. J. Physiol. Lond., temporal retina 216: 39-68. Blakemore, C. 1969 Binocular depth discrimi- In the context of an analysis of the naso- nation and the nasotemporal division. J. temporal division of the cat's retina (Stone, Physiol. Lond., 205: 471-497. '66) evidence was presented that ganglion Boycott, B. B., and J. E. Dowling 1969 Or- cells in the area of retina temporal to the ganization of the retina: micros- area centralis project to both ipsi- and con- copy. Phil. Trans. Roy. SOC.B., 255: 109-184. tralateral optic tracts, and the suggestion Fukada, Y. 1971 Receptive field organisation of cat optic nerve fibres with special reference was made that the axons of these contra- to conduction velocity. Vision Res., 11: 209- laterally-projecting cells terminate in mid- 226. brain visual centres. The presence of con- Gartner, S. 1951 Ocular pathology in the tralaterally projecting cells in temporal . Amer. J. Ophthal., 34: retina of the cat has been confirmed 593-596. (Laties and Sprague, '66; Ogawa et al., Hoffmann, K.-P. 1970 Retinotopische Bezie- '69; Fukada, '71; Sanderson and Sherman, hungen und Struktur rezeptiver Felder im '71), and their projection to the Tectum opticum und Practectum der Katze. has been demonstrated (Laties and Z. vergl. Physiologie, 67: 26-57. Kupfer, C. 1963 degen- Sprague, '66; Hoffmann, '69). Sanderson eration following chiasmal lesions in man. and Sherman ('71) have however pre- Arch. Ophthal., 70: 256-260. sented evidence that some of these cells Lane, R. €I,, J. M. Allman and J. H. Kaas project to the , specifically to the 1971a Representation of the visual field in medial interlaminar nucleus and to layer the oi the grey squirrel (Sciurus carolinensis ) and the tree shrew B of the lateral geniculate nucleus. (Tupaia glis). Brain Res., 26: 277-292. Contralaterally-projecting ganglion cells Lane, R. H., J. M. Allman, J. H. Kaas, F. M. appear to be present in the temporal retina Miezin and C. N. Woolsey 1971b The repre- of other non-primate mammals. For ex- sentation of the visual field in the superior ample in both the grey squirrel and the colliculus of the owl monkey (Aotus trivir- tree shrew temporal retina projects to the gatus). Fed. Proc., 36: 615P. Laties, A. M., and J. M. Sprague 1966 The contralateral superior colliculus (Lane projection of optic fibers to the visual centers et al., '71a). Evidence presented in this in the cat. J. Comp. Neur., 127: 35-70. report suggests that such cells are not Linskz, A. 1952 Physiology of the Eye. Vol. 2. present in the monkey; i.e. all the gan- Vision. Grune and Stratton, New York, 346 pp. glion cells located temporal to the strip Nikara, T., P. 0. Bishop and J. D. Pettigrew of overlap appear to project to the ipsi- 1968 Analysis of retinal correspondence by lateral optic tract. This is in accord with studying receptive fields of single units in cat Lane et al.'s ('71b) evidence that in the striate cortex. Exp. Brain Res., 6: 353-372. Ogawa, T., Y. Imazawa and S. Chu 1969 Elec- owl monkey (in to the grey squir- trophysiological tracings of intraretinal optic rel and tree shrew) the superior colliculus nerve fibers in the cat. Tohoku J. exp. Med., does not receive fibres from ganglion cells 98: 215-222. in the temporal area of the contralateral Ogle, K. N. 1962 Spatial localisation through retina. It may prove to be the case that an binocular vision. In: The Eye. Vol. 4. H. Davson, exclusively ipsilateral projection of tempo- ed. Academic Press, N. Y. and Lond., pp. 350- 407. ral retina is to be found only in . Polyak, S. 1957 The . H. Kluver, ed. Univ. of Chicago Press, Chicago, LITERATURE CITED Illinois. Borlow, €1. B., C. Blakemore and J. D. Pettigrew Sanderson, K. J., and S. M. Sherman. 1971 1967 The neural mechanism of binocular Nasotempord overlap in visual field projected 348 J. STONE, J. LEICESTER AND S. M. SHERMAN

to the lateral geniculate nucleus in the cat. nodal distance and retinal landmarks. Vision J. Neurophysiol., 34: 453466. Res., 3: 289-314. Stone, J. 1965 A quantitative analysis of the van Buren, J. M. 1963 The retinal ganglion distribution of ganglion cells in the cat’s retina. cell layer. Charles C Thomas, Springfield, J. Comp. Neur., 134: 337-352. Illinois, p. 42 ff. 1966 The naso-temporal division cf Williams, D., and M. M. Gassel 1962 Visual the cat’s retina. J. Comp. Neur., 136: 585-600. function in patients with homonymous hemi- Vakkur, G. J., P. 0. Bishop and W. Kozak 1963 anopia: Part I, The visual fields. Brain, 85: Visual in the cat, including posterior 175-250.