Cytochrome-oxidase blobs in cat primary visual cortex

The Journal Cytochrome-Oxidase Kathryn M. Murphy,’ David of Neuroscience, June 1995, 15(6): 4196-4208 Blobs in Cat Primary Visual Cortex G. Jones,2,a and Richard C. Van Sluyters3 ‘McMaster University, Neural Organization and Plasticity Laboratory, Department of Psychology, Hamilton, Ontario L8S 4K1, Canada, 2McGill University, Department of Electrical Engineering, Montreal, PQ H3A 2A7, Canada, and 3University of California, School of Optometry, Berkeley, California 94720-2020 Cytochrome-oxidase blobs are central to two of the most influential ideas in contemporary visual neurosciencecortical modularity and parallel processing pathways. In particular, the regular 2D array of cytochrome-oxidase-rich blobs in primate visual cortex is arguably the most compelling evidence for cortical modularity and has been hypothesized to mark a separate processing stream through the visual cortex. Although previously a variety of mammals have been studied, blobs have only been demonstrated in the visual cortex of primates, which has led to the conclusion that blobs represent a primate-specific feature of visual cortical organization. Here we demonstrate the presence of cytochrome-oxidase blobs in a nonprimate species. Throughout the full tangential extent of layers IIIll in cat visual cortex the cytochrome-oxidase staining pattern is distinctly patchy, with the darkly stained blobs forming a regular 2D array. In addition, the blobs in cat visual cortex are functionally related to the underlying ocular dominance columns. The presence of cytochrome-oxidase blobs in the cat clearly demonstrates that they no longer can be considered a primate-specific feature of visual cortical organization. [Key words: cytochrome-oxiclase, blobs, cat, visual cortex, modules, ocular dominance] Features of the mature visual cortex are distributed in a periodic fashion across its tangential extent forming the familiar modular structure of this cortical area (Hubel and Wiesel, 1962, 1974). A relatively simple histochemical staining technique for the mitochondrial enzyme cytochrome oxidase (CO) (Wong-Riley, 1979) produces a notably regular pattern of darkly stained patches in the upper layers of primate visual cortex (Horton and Hubel, 1981; Horton, 1984). The darker CO-stained patches are called blobs, and the lighter intervening regions, interblobs (Livingstone and Hubel, 1984a). The 2D pattern of CO blobs has become the most easily visualized neuroanatomical correlate of, and arguably the most compelling evidence for, cortical moduReceived Aug. 17, 1994; revised Oct. 21, 1994; accepted Dec. 7, 1994. This work was supported by NSERC grants (URF0124132, EQP0043821, lCR0100439) to K.M.M., NIH grant (EY02193) to R.C.V.S., and NIH CORE Facilities Support Grant (EY03176). K.M.M. is an NSERC University Research Fellow and an Alfred P. Sloan Research Fellow. We are especially indebted to Drs. Michael P Stryker, Steven W. Zucker, and Jonathan Horton for their helpful discussion of these data. Correspondence should be addressed to Dr. Kathryn M. Murphy, McMaster University, Neural Organization and Plasticity Laboratory, Department of Psychology, 1280 Main Street West, Hamilton, Ontario L8S 4K1, Canada. “Present address: McMaster University, Department of Computer Science, 1280 Main Street West, Hamilton, Ontario LSS 4K1, Canada. Copyright 0 1995 Society for Neuroscience 0270-6474/95/154196-l 3$05.00/O larity in primate visual cortex. The distribution of CO blobs has provided a robust, albeit opportunistic, anatomical marker with which to correlate the distribution of physiological properties in the visual cortex, the array of local cortical circuits, and the arrangement of corticocortical connections. These studies have allowed the formation of links between physiological properties and the underlying anatomy of both the modular organization of the primary visual cortex and the streaming of visual information into parallel processing pathways (Zeki, 1975; Livingstone and Hubel, 1983, 1984a; DeYoe and Van Essen, 1985, 1988; Shipp and Zeki, 1985; Van Essen et al., 1992). In the monkey, blobs lie at evenly spaced intervals along ocular dominance columns (Fitzpatrick and Diamond, 1980; Hendrickson et al., 1981; Horton, 1984) and are linked by a latticework of blob-specific horizontal connections (Livingstone and Hubel, 1984b). Cells within blobs tend to be less binocular and less selective for orientation, but more color selective, than cells in interblob regions (Livingstone and Hubel, 1984a), and projections from blob or interblob regions to higher cortical areas exhibit a certain degree of segregation (Zeki, 1975; Livingstone and Hubel, 1983, 1984a, 1987a,b). Hence, in addition to being an impressive anatomical demonstration of cortical modularity, the CO blobs have proven to be a key feature for understanding the parcelling and streaming of functions within the primate visual system. Although previously a variety of mammals have been studied (e.g., Wong-Riley, 1979; Horton, 1984; Price, 1985), blobs have only been demonstrated in the visual cortex of primates, which has led to the widely accepted notion that blobs represent a primate-specific feature of visual cortical organization. The lack of evidence for CO blobs in other mammals resulted in primates becoming the animal model for studies of parallel processing streams and frustrated attempts to formulate general models for the functional organization of modules in the mammalian visual system. For example, the pattern of CO staining in the cat visual cortex has been described as uniform within the supragranular layers (e.g., Wong-Riley, 1979; Horton, 1984; Price, 1985), thereby raising the possibility of a significant interspecies difference, even at this relatively early stage in the visual pathway, in the other higher mammal commonly studied in visual neuroscience. Thus, although other features such as ocular dominance columns appear to arise from similar constraints in cats and monkeys (Jones et al., 1991), the apparent lack of blobs in the cat became an impediment for integrating information gleaned from experiments using the cat model into an overall understanding of the significance of cortical modularity for visual perception. In our previous studies of the anatomical organization of oc- The ular dominancein the cat visual cortex, we came to appreciate the insightsthat can be gained when the overall organization of this pattern is examinedin large, tangential sectionscut from an unfolded and flattened cerebral hemisphere(Anderson et al., 1988; Murphy et al., 1989). For example, viewed in this way, the ocular dominancepattern is easily recognized as a two-dimensionalnetwork of branching, beadedbandsthat are not oriented along any particular axis. In our experience, examination of sectionscut tangential to the surface of the unfolded and flattened cortex provides the most sensitive technique for resolving virtually any anatomicalpattern that is arrayed parallel to the cortical laminae.This is especially true when the histological image of the anatomicalpattern is of low contrast. For this reason,along with the possibility that blobs might be spatially related to the underlying pattern of ocular dominancecolumns in the cat, we were encouragedto reexaminethis issuein tangentialsectionscut from unfolded and flattened visual cortex. Here, we demonstratethat cytochrome-oxidase blobs are, indeed, a feature of primary visual cortex in nonprimatespecies. Preliminary reports of this discovery have included someof thesedata (Murphy et al., 1990, 1991a,b). Materials and Methods Animuls and surgical procedures. A total of 10 adult cats were used to study the organization of CO blobs in the visual cortex. The cats were placed into one of four experimental groups; no manipulations (n = 3), unilateral eye injection with wheatgerm agglutinin conjugated to horseradish peroxidase (WGA-HRP) (IZ = 3), unilateral retina lesion (n = 2), and unilateral retinal lesion plus WGA-HRP injection into the fellow eye (n = 2). The complete pattern of cortical ocular dominance was revealed following unilateral intraocular injections of the anterograde transneuronal tracer WGA-HRP (Sigma, St. Louis, MO) (5% in saline), using previously described procedures (Anderson et al., 1988). To further examine the relationship of the CO blobs to cortical ocular dominance columns, a complete retinal lesion of one eye was performed. Aseptic surgical techniques, gaseous anesthetic (0.5-5.0% halothane in 67% N,0/33% 0,), and postsurgical analgesics were used for the retinal lesion procedure. A circumlimbial incision of the cornea was made, exposing the iris and lens. Using light suction, the lens was extracted through the pupil, taking care to not tear the iris or ciliary body. The vitreous humor was aspirated, then the whole retina was swabbed free from the eyecup. Saline-and-thrombin-soaked Gelfoam was used to pack the eyecup tightly and the circumlimbial incision was sutured closed. A broad spectrum ophthalmic antibiotic was applied to the front of the closed eyecup, then the lid margins were sutured together for about 7 d to protect the ocular incision while it healed. Postlesion survival periods ranged from 7 to 90 d. Histology. Animals were euthanized with a lethal injection of Nembutal (150 mg/kg, i.p.) then perfused transcardially with saline (1 -I .5 1 at 45°C) followed by 4 min of fixation with 2% glutaraldehyde in 0.1 M phosphate buffer (4°C). This perfusion protocol was optimized for the process of unfolding and flattening the cortex; glutaraldehyde fixes the gray matter while leaving the white matter relatively unfixed. The brain was removed from the cranium and the cerebral hemispheres were resected for unfolding and flattening (Olavarria and Van Sluyters, 1985). The block of brain containing the dLGNs was placed in 30% sucrose, 2% glutaraldehyde in 0.1 M phosphate buffer and refrigerated (4°C). During the unfolding and flattening of the cortex the hemispheres were kept wet with 2% glutaraldehyde while the white matter was carefully excised. After removal of the white matter, the intrinsic curvature in the sheet of gray matter was relieved by making a medial and a lateral cut, thereby allowing the visual cortex to lie flat. It is important to note that at this stage in the unfolding and flattening procedure the cortical tissue was well fixed, and these cuts in the tissue were essential to prevent tearing or possible distortion of the visual cortex during flattening. Once flattened, the hemispheres were postfixed with 2% glutaraldehyde, 30% sucrose in phosphate buffer (4°C) for up to 6 hr, then transferred to 30% sucrose in phosphate buffer (4°C) and stored overnight. The following day the unfolded and Battened cortices were cut tangential to the pial surface (50 km) on a freezing microtome and the sections were Journal of Neuroscience, June 1995, 75(6) 4197 collected in phosphate buffer. For those cases where both CO and labeled ocular dominance columns were studied, all the supragranular sections to a depth of 550 p,rn were stained for CO, then alternate sections were either stained for CO, or reacted with TMB to reveal the labeled ocular dominance columns. The sections for CO staining were mounted onto gelatin-coated glass slides, allowed to air dry, and then reacted (5-7 hr incubation at 40”) following the standard CO histochemical procedure used by Horton (1984). All sections from a hemisphere were reacted together for the same incubation period. The dLGNs were sectioned coronally on a freezing microtome (50 km). mounted onto slides, and air dried, then reacted for CO (2-4 hr incubation). Stained slides were dehydrated and defatted, coverslipped with DPX (Aldrich, Milwaukee, WI), and allowed to dry before photographing. Sections reacted to reveal the labeled pattern of ocular dominance columns were collected in phosphate buffer, washed in three changes of buffer, then reacted with the chromogen tetramethylbenzidine (Sigma, St. Louis) using the Mesulum histochemical procedure as modified by Gibson et al. (1984). This modification keeps the A and B parts separate and continuously filters the B solution, thereby minimizing the amount of precipitate and resulting in very high sensitivity for revealing transneuronally transported WGA-HRF? Once any bluish precipitate was washed out in the postreaction rinse, the sections were mounted onto gelatin-coated glass slides and air dried. The sections were dehydrated and defatted, then coverslipped using DPX. Tangential analysis. The relationship between the CO blobs and ocular dominance columns was studied in two groups of animals: one in which ocular dominance columns were visualized following an eye injection of WGA-HRP, and a second where one eye was injected with WGA-HRP while retinal activity was silenced in the fellow eye by lesioning the whole retina. In both of these groups, the tangential relationship of the blobs in layers IIll was compared with that of the ocular dominance pattern from layer IV. Using radial blood vessels to manually align sections through layers II/III and IV, a region with distinct blobs was analyzed. These successive tangential sections through layers II/III and IV were input to a computer system (SUN4) by imaging with a color CCD camera (Sonv, XC-7 11) connected to a large-format macrophotography system and frame-grabber board (Parallax Graphics, XVideo/RGB). The low magnification (0.7-5.0X), good color resolution of the imaging system, and ability to store and view multiple aligned raw images, were helpful in locating the CO blobs. First, blobs were visualized and marked with a cross using the computer system to view the aligned raw images from all supragranular CO-stained sections in rapid succession, in a manner akin to focusing up and down through the sections. Next the resulting pattern of crosses was overlaid onto aligned serial images of the ocular dominance pattern representing the full thickness of layer IV; the crosses were then color coded such that crosses coincident with the labeled ocular dominance pattern were coded red, and crosses coincident with the unlabeled ocular dominance regions were coded green. During the analysis, the positions of the blobs were localized without any prior knowledge of the underlying ocular dominance pattern, and this was done independently by two observers. This analysis was performed on rectangular regions within area 17, varying in size from 4.5 X 3 mm to 10 X 7 mm. The technique for serially viewing all of the CO-stained sections on the computer aids in the analysis of these data in two ways. First, the third dimension is included, thereby eliminating the problem of missing blobs in a single 2D section because regions had dipped out of the plane of section. Second, viewing successive aligned sections improves the signal-tonoise, since the spatial location of the blobs is consistent through many images, while nonspecific variations in staining show no spatial correlation across sections. Thus, the apparent contrast of the blobs is improved. In addition, a second analysis was performed with aligned single sections, one section from layers II/III stained for CO with obvious blobs, and the other section from layer IV reacted to reveal the pattern of ocular dominance following an eye injection. Each image was thresholded at midluminance to produce binarized images; the binarized blob and ocular dominance images were added to generate a color-coded image of the relationship between blobs and ocular dominance. Results When single tangential sectionsthrough the posterior neocortex of the cat are stainedfor CO, the ovate, two-dimensionalshape of areas17 and 18 standsout as a region that generally stains 4199 Murphy et al. * CO Blobs in Cat Area 17 layers from the unfolded and flattened posterior (4.2 X 3 cu., au P..L ..wk), througl. ,,,,,,,,.,,lar Figure 1. A, A neocortex of the cat, stained for cytochrome oxidase. The complete 2D extent of area 17 is visible in this section as the darker stained The Journal of Neuroscience, June 1995, 75(6) 4199 Figure 2. The normally reared LU UKIKO 1x1 more darkly than the surrounding cortical areas. Closer inspection of the staining pattern reveals that within area 17 of the cat cortex the staining pattern is nonuniform in certain sections. Throughout the full extent of area 17 the staining pattern in tangential sections passing through layers II-III is distinctly patchy-CO blobs are, indeed, present in supragranular layers (Fig. 1A). The nature of these patches of darker CO staining is even more apparent at slightly higher magnification (Fig. 1B). The blobs were generally somewhat more distinct in the region of area 17 in which the central part of the visual field is represented (see Fig. 1B) than in the region representing the periphery of the visual field. This tangential variation in the qualitative appearance of the blobs was a consistent feature of all hemispheres studied and resembles tangential variations present in the cat ocular dominance pattern (Anderson et al., 1988; Murphy et al., 1989). Further analyses are confined to the more central visual field representation (< 10”) within area 17 .where both the blobs and ocular dominance patterns are most easily distinguished. Blobs were observed in the primary visual cortex in both hemispheres of all cats studied. Figure 2 shows the appearance of the blobs observed in each of the six normal cats studied. It is apparent from comparison of the pattern of CO staining shown in the panels of Figure 2 that the shape, size, and spacing of the blobs appears similar in all of the cats studied. In primate cortex, CO staining is nonuniform in a number of extra-striate cortical areas, most notably V2, which has an arrangement of thick and thin darkly stained stripes (Horton, 1984). In our initial examinations of cat area 18 we have not observed any consistent pat- :ach of the six terning of the staining in this area. This may reflect the fact that unlike the primate, area 18 in the cat receives some direct input from LGN Y-cells, or it may simply be that the incubation times used to achieve optimal staining of area 17 are not appropriate for staining area 18. It is apparent in the CO-stained tangential sections through areas 17 and 18 (see Fig. 1A) that, as has been reported using coronal sections (Price, 1985), there is no abrupt change in CO staining at the area 17/18 border and that the dramatic change in the density of staining occurs more lateral at the area 18/l 9 border. A patchy pattern of CO staining in layer IV of cat visual cortex had been reported previously only following disruption of normal binocular visual experience (Wong-Riley, 1979; Wong-Riley and Carroll, 1984). If an abnormality associated with one eye resulted in the patchy pattern of CO staining in the cortex observed in the present study, then it also should have resulted in reduced staining of the layers of the dorsal lateral geniculate nucleus (dLGN) that receive their input from that eye (Wong-Riley and Riley, 1983; Kageyama and Wong-Riley, 1986). Examination of dLGN sections reacted for CO revealed that the staining was of comparable density in the right- and left-eye layers of the normally reared cats (Fig. 3A). This is contrasted with the staining pattern found following monocular retinal lesion where there was a reduction in the density of staining in the dLGN laminae subserving the lesioned eye (layer Al in Fig. 3B). Having established the presence of blobs in the cat visual cortex, some of their anatomical characteristics were examined. The radial extent of the blobs was assessed by using blood vessel t Within area 17, the cytochrome-oxidase ovate region that extends across almost the full width of the photomicrograph (delineated by filled arrows). staining is noticeably patchy, and these darker patches or blobs are apparent throughout virtually the entire extent of area 17 in this single supragranular section. Some variation in the density of the background staining is apparent where there is a change in the laminar plane of section. In this section, anterior is to the right and medial toward the top. The two cuts, one medial (open arrow) and the other lateral, were made to relieve the intrinsic curvature of the cortex so that the tissue could lie flat. The scale bar represents 1 cm. B, A higher magnification view of the blobs from the area centralis representation of the section shown above (open arrow indicates the “V-shaped” medial relieving cut). The blobs are darker patches of cytochrome-oxidase staining that are distinct from the interblob zones. The qualitative nature of the blobs in cat visual cortex appears similar to that observed in macaque monkey visual cortex. Scale bar, 2 mm. 4200 Murphy et al. l CO Blobs in Cat Area 17 Figure 3. A, The pattern of CO staining in the dLGN density. B, CO-stained section from the dLGN ipsilateral lamina that receives inout from the lesioned retina (Al dorsal is to the top andmedial to the right. Scale bar, 1 of a normally reared cat. The A and Al laminae are darkly stained and of comparable to an eye with a complete retinal lesion. The density of staining is notably lighter in the laver) than the layer that receives input from the intact eye (A layer). In these sections, m&. ’ landmarksto align serial tangential sectionsand following the staining pattern for individual blobs through adjacent sections. Typically, blobs are first faintly visible at a depth of about 250300 pm below the pial surface (Fig. 4A). The blobs become progressively darker, reaching their maximum density at about 450-500 Frn below the surface (Fig. 4C). Beneath this depth, the blobs rapidly mergetoward a uniformly densestainingpattern within layer IV, so that by about 550-600 pm below the pial surface(Fig. 4E) they are no longer visible. Often the blob pattern is only easily discernible where the contrast between blob and interblob regions is the highest (i.e., about 450 p,rn below the pial surface). This latter finding underscoresthe importance of carefully unfolding and flattening the visual cortex in order to obtain tangential sectionswith large regionsthat pass through the portion of layers II-III where blobs are most easily discriminated.Compiling the radial changesobservedfrom several animalsreveals that blobs are largely confined to layers IIIII of cat visual cortex and that their total radial extent is approximately 250-300 pm. We extended the radial analysis of the blobs in the cat to include comparisonwith the location of the geniculocortical terminations,labeledfollowing an eye injection, in comparableseries of tangential sectionsfrom an earlier study (Anderson et al., 1988). Since all sectionsin this and the earlier study were from unfolded and flattened cortices cut tangentially at a thicknessof 50 pm, it was possibleto compareacrossanimalsand begin to analyze the depth below the pial surface where either blobs or geniculocortical inputs are located. AS describedabove, the radial location of the blobs is from a depth of about 250-500 p,rn and geniculocortical afferents were seento occupy the region extending from about 450 to 850 km below the pial surface (Anderson et al., 1988). Clearly, most of the radial extent of the blobs must be superficial to the geniculocortical afferents, and the laminar positionswhere each of thesepatterns is most distinct are not coincident. However, at their deepestextent the blobs may overlap the zone of geniculocorticalinput (LeVay and Gilbert, 1976; Leventhal, 1979). It shouldbe recognized that the presenceof CO blobs in the monkey was a mere curiosity until they were linked with other featuresof cortical organization. One of the first relationshipsto be examinedin monkeyswas that betweenthe blobs and cortical ocular dominancecolumns, where it was shown that blobs are both spatially correlated with ocular dominance columns and functionally eye specific (Fitzpatrick and Diamond, 1980; Hendrickson et al., 1981; Horton, 1984). In the present study, the spatialrelationshipwasexaminedin an additional threenormally reared adult cats in which the tangential patterns of both CO blobs and ocular dominancecolumnsin area 17 were visualized within the samehemisphere.An example of an analysis performed to examine whether there is a simplespatialrelationship betweenthe position of the CO blobs and the ocular dominance patternsis shown in Figure 5, where the tangential location of blobs in all sectionsthrough layers II/III were marked (Fig. 5A; seeMaterials and Methods for details) and aligned using radial blood vessels with the underlying WGA-HRP labeled ocular dominancepattern from layer IV. The locations of the blobs were then color coded to reflect whether a blob was associated with a labeled (red crosses)or an unlabeled(green crosses)ocular dominance column. In the color-coded map (Fig. 5B), it appearsthat there is a relatively even mixture of blobs that are associatedwith either the ipsilateral (labeled) eye’s ocular dominancecolumns(red crosses),or the presumedcontralateral(unlabeled)eye’scolumns(greencrosses).The presenceof both red and green crossesin this analysisindicatesthat the blobs do not simply reflect the pattern of one eye’s inputs. The spatial locations of thesered and greencrosses,however, are not positioned solely above the centers of the underlying ocular dominance columns.Thus, an additional analysisof this spatialrelationship was done to assessthe area1extent of blobs with respectto the pattern of ocular dominance.From this analysis we could visualize whether blobs located near an ocular dominanceboundary are encompassed by or straddleocular dominancedomains. Regions from single supragranularsectionswith distinct blobs (Fig. 6A, red blobs) and a layer IV section reactedto reveal the pattern of inputs from one eye (Fig. 6B, green labeled ocular dominance)were aligned, binarized, color coded, and then added together (Fig. 6C; see Materials and Methods for details). Clearly, from this analysis (Fig. 6C) some blobs are situated largely within the unlabeled (largely red blobs) regions of the ocular dominancepattern, while others are encompassed by the The Journal of Neuroscience, June 1995, 75(6) 4201 labeled(yellow blobs) zones, and somestraddlethe labeledand unlabeleddomains (red and yellow). The results of these two analysesindicate that in the cat the spatialrelationshipbetween blobs and ocular dominance does not follow the same simple pattern observed in primates (see Discussion). However, the finding in normal cats that someblobs do overlie either the labeled or unlabeledaspectsof the underlying ocular dominance pattern leavesopenthe possibility that blobsmay befunctionally eye specific. At this point, the relationshipbetween ocular dominanceand blobs had only been examined as a simple spatial correlation from which it is tenuous to either draw or eliminate possible functional relationships.Therefore, we challengedthe hypothesisthat blobs may be functionally eye specific by using one of the techniquesemployed in previous primate studiesto address this issue(e.g., Horton, 1984), that of silencingthe activity from one eye. The notion of a functional relationship was tested by examining the arrangementof blobs with ocular dominancecolumns in four cats that were monocularly enucleatedas adults reduces CO staining in the layers of (Fig. 7). This manipulation the dLGN (Fig. 3B) and in the cortical ocular dominancecolumns correspondingto the enucleatedeye. If the blobs are not eye specific, then one would expect their tangentialarrangement to be unchangedby this manipulation, and the analysisof their relationship to ocular dominancecolumns would still have red and greencrosses.Conversely, if the blobs are eye specific then, as has been demonstratedin the monkey (Horton, 1984), there should be a significant reduction in staining of blobs overlying the enucleatedeye’s ocular dominancecolumns,possibly to the point where those blobs are no longer discernible,and the analysis would result in crossesof only one color. To ensurethat a reduction in contrast of some blobs did not bias this analysis against finding blobs over unlabeled Figure 4. Rat changes m the pattern ot CO stammg through layers II/III (A-C) and into upper layer IV (D-E). The sections were aligned, using the radial blood vessels, to allow for comparison of the CO staining pattern in these layers. The pattern of CO staining (A) 250 pm, (B) 350 pm, and (C) 450 urn, below the pial surface is consistently patchy. Using the radial blood vessels as landmarks to align sections, it is clear that the dark patches of CO staining are in register radially and form a columnar structure. The subsequent panels show (D) the pattern of CO staining at the transition between layers II/III and layer IV (500 pm regions of the ocular dom- inance pattern, care was taken to mark every region of denser CO staining that may possibly representa blob. A color-coded map (Fig. 7B) of the relationship between the 2D arrangement of CO blobs and ocular dominancecolumns followmg monocular enucleation showsthat blobs now are almost exclusively associatedwith the remaining eye’s ocular dominance columns-there are only red crossesin this analysis.Further analysis of the relationship between blobs and ocular dominance following enucleation reveals that blobs are virtually encompassed(yellow regions)by the underlying ocular dominancepattern (Fig. S), suchthat very little red remainswhen the binarized blobs and ocular dominancepatterns are added. This contrasts with the arrangementfound in nonlesionedcats, where many blobs overlie the unlabeledregionsof the ocular dominancepattern (seeFigs. 5B, 6C). Following a completeretinal lesion,we could find no blobsthat appearto be above the unlabeledregions of the ocular dominancepattern, the presumedmonocular aspects of the lesionedeye’s ocular dominancecolumns.The results of these two analyses(Figs. 7B and SC) solidifies the argument for a functional relationship between blobs and ocular dominancein the cat. Furthermore, the remaining CO blobs in layer II/III were patchy (seeFigs. 7A, 8A) and so it was not the casethat the pattern of CO staining in the supragranularlayers merely reflected the complete pattern of geniculocortical affer- below the oial surface), and (E) laver IV (5.50 nm below the oial surface) where the CO staining pattern becomes dark and relati;ely uniform. Scale bar, 1 mm. 4202 Murphy et al. * CO Blobs in Cat Area 17 Figure 5 (left). Using radial blood vessels to align a series of CO-stained sections with those reacted to reveal the pattern of ocular dominance columns, a region with distinct blobs was overlaid onto the underlying ocular dominance pattern. A, In this 10 by 7 mm portion of a tangential section from layers II-III (450 brn below the pial surface) CXLKKY mark the location of blobs. The blobs were located by comparison with the adjacent cytochrome-oxidase stained sections and in this example the range of center-to-center spacing is approximately 350-1000 km. B, This image shows the relationship that is revealed by overlaying the crosses onto the underlying ocular dominance pattern revealed following intraocular injection of the transneuronal tracer WGA-HRP. In this dark-field view of a section from a depth of 750 km (i.e., 300 pm below that in B), the light bands represent the arrangement of the geniculocortical inputs from the injected ipsilateral eye. Where the blobs are coincident with the labeled ocular dominance columns, the crosses are color coded reel, and where they are coincident with the unlabeled ocular dominance regions the crosses are coded green. Similar results were obtained when the hemisphere contralateral to an injected eye was examined. Scale bar, 2 mm. Fig~tre ocular ocular shown 6 (vight). dominance dominance in yellow, Visualization of the relationship between aligned regions of (A) CO blobs (red) in single sections through layers II/III, and (B) (green) in layer IV revealed following an eye injection with WGA-HRP in a normally reared cat. Images of the CO staining and were thresholded at midluminance then added to generate a color-coded image. C, CO blobs over the labeled eye’s inputs are blobs over the unlabeled regions are shown in red. Scale bar, 1 mm. ents subserving the surviving eye (see Figs. 7B, 8B). Thus, the functional relationship between blobs and ocular dominance columns in the cat seems similar to that found in the monkey; in both species, blobs exhibit eye specificity. Parenthetically, no previous study in the cat had demonstrated that the pattern of CO staining observed in layer IV following disruption of retinal activity reflects the remaining eye’s ocular dominance columns. In this study, the combination of a retinal lesion, CO staining, and an eye injection of a transneuronal tracer to visualize the pattern of ocular dominance in layer IV, answered this question. Images of adjacent sections from layer IV, either stained for CO or reacted to reveal the pattern of the surviving eye’s inputs, were added together to determine the relationship between these patterns. A negative contrast image of CO staining (dense CO staining 1s white) was added with the dark-field image of the ocular dominance pattern. With this analysis, it is straight forward to visualize whether the labeled ocular dominance and dense CO-staining patterns are in register or complementary. In the situation where these two patterns are in register, the resulting image would appear very similar to each of the component images. On the other hand, if the ocular dominance and CO images are complementary, then the resulting image would be uniformly white. The visual cortices examined in this experiment showed good agreement in layer IV between the pattern of CO staining (Fig. 9A) and the pattern of ocular dominance columns labeled following injection of the surviving eye (Fig. 9B). Since the anatomical features present in CO staining (Fig. 9A) and labeled ocular dominance (Fig. 9B) persist in The Journal of Neuroscience, June 1995, 15(6) 4203 Figure 7 (left). The relationship between CO blobs and the underlying ocular dominance pattern is shown following 2 weeks survival after monocular enucleation. A, Cytochrome-oxidase blobs are apparent in this 7 by 5 mm portion of a tangential section through layers II-III. B, By comparison with adjacent CO-stained sections, the blobs in A are marked with crosses. The arrows indicate some of the radial blood vessels used for alignment. B, A color-coded map of the relationship of blobs to the underlying ocular dominance columns revealed following an eye injection of WGA-HRP, prepared according to the methods described in Figure 5. The light bands represent the geniculocortical inputs from the remaining eye. As indicated by the red crosses, after monocular enucleation, all of the blobs in layers II-III overlie the remaining eye’s ocular dominance columns. Scale bar, 2 mm. Figure 8 (right). Following complete monocular retinal lesion, the arrangements of (A) CO blobs (red) and (B) labeled ocular dommance pattern (green) of the surviving eye were compared. Images were prepared following the method described in Figure 6. C, In the color-coded map, CO blobs over the surviving eye’s inputs are shown in yellow while the small amount of CO staining over the unlabeled regions is shown in red. Scale bar, I mm. the summed image (Fig. 9C), this confirms that the dense CO staining in layer IV after a total retinal lesion is in register with the regions subserved by the remaining eye’s geniculocortical inputs (Fig. 9C). Discussion In this study we report patchy, nonuniform CO staining in the upper layers of cat visual cortex. This patchy pattern of dark CO staining is within the supragranular layers and bears a qualitative similarity to demonstrations of CO blobs in macaque visual cortex. In addition, the dark patches of CO staining in cat visual cortex appear to be functionally related to the modular organization of this area. Thus, based on similarities in both their appearance and their relationship to other features of the visual cortex, it seems prudent to apply the accepted nomenclature from the primate to describe the patterns of CO staining observed in these two species. Hence, we have referred to the dark patches of CO staining in cat visual cortex as blobs. This discovery lags by more than a decade the initial reports of CO blobs in primate Vl (Horton and Hubel, 198l), and subsequent attempts by numerous other groups (e.g., Wong-Riley, 1979; Horton, 1984; Price, 1985) to demonstrate the presence of CO blobs in nonprimate species. At first glance, the present demonstration of CO blobs in the visual cortex of a nonprimate species may seem somewhat surprising, since until now it had been widely accepted that CO blobs represent a primate-specific feature of the visual cortex. However, using the standard CO histochemistry from Horton (1984) on sections from the unfolded and flattened visual cortex, blobs have proven to be a robust feature of layers II/III in adult cat visual cortex. The major technical difference in the present approach is that CO staining was examined throughout the complete tangential extent of area 17, in sections from unfolded and flattened cat cortex, whereas previous attempts had examined 4204 Murphy et al. l CO Blobs in Cat Area 17 Figure 9. The relationship between CO staining and labeled ocular dominance patterns in adjacent sections from layer IV following monocular retinal lesion. A, A negative image of CO staining in layer IV after lesioning one eye’s retina (white represents dense CO staining). B, Dark-field image of the ocular dominance pattern following injection of WGA-HRP into the surviving eye (white represents labeled inputs). C, The aligned images from A and B were added together to demonstrate that there is good correspondence between the regions of dense CO staining in layer IV after monocular retinal lesion and the surviving eye’s inputs. The features of the pattern in C resemble the anatomical features present in both A and B. radially cut sections.The finding that blobs are readily apparent in sectionscut tangential to the pial surface in the cat affirms that this approachis more sensitive for revealing virtually any anatomicalpattern arrayed parallel to the cortical laminae, especially if the pattern is of low contrast. To our knowledge, no other histochemical staining method has revealed such a widespread,regular, patchy pattern that is an intrinsic feature of the supragranularlayers in visual cortex of normally.reared adult cats. Although, a recent report hasdescribed a patchy distribution of a molecularly distinct classof pyramidal neuronsin layers II/III of adult cat visual cortex (Mareschalet al., 1994), and others have noted during development transiently patchy distributions of various markers,somein layers II/III, in the kitten visual cortex (Schoenet al., 1990; Dyck et al., 1993; Dyck and Cynader, 1993a), as well as patchy expressionof the proto-oncogeneC--OS(Beaver et al., 1992). Following our initial reports of CO blobs in adult cats (Murphy et al., 1990, 1991a,b),Dyck and Cynader (1993b) have observed patchy CO staining in the visual cortex of the kitten. Surprisingly, they have reported that at postnatalday 50 CO staining is patchy in layer IV, in contrast to the presentfindings in adult cats, and previous data from primates (e.g., Horton, 1984), which show that CO staining is relatively uniform in layer IV and most patchy supragranularly. Prior to the present study, nonuniform CO staining in adult cat visual cortex had beendemonstratedin layer IV only following disruption of binocular vision (Wong-Riley, 1979; WongRiley and Carroll, 1984). There are several lines of reasoningto indicate that the variegated pattern of CO staining we find in layers II-III is not an inadvertent demonstrationof the underlying ocular dominancepattern. First, all six cats studied were reared with normal binocular visual experience, and yet each clearly demonstratedblobs in both visual cortices. Furthermore, blobs were presentin sectionsthrough layers II-III, whereasthe stainingpattern within layer IV was uniform (Fig. 4). Finally, if an abnormality associatedwith one eye had disruptedthe pattern of cytochrome-oxidasestaining in the cortex, then it also should have resulted in reduced staining of the layers of the dorsal lateral geniculate nucleus (dLGN) that receive their input from that eye (Wong-Riley and Riley, 1983; Kageyama and WongRiley, 1986). Examination of dLGN sectionsreacted for cytochrome oxidase revealed that the staining was of comparable density in the right- and left-eye layers of the normally reared cats (Fig. 3A), as comparedwith the CO staining of the dLGN layers following a monocular retinal lesion (Fig. 3B). All of these points indicate that there was no hidden abnormality in the primary visual pathway of thesecats that could have caused a patchy pattern of cytochrome-oxidasestaining in layers II-III. Evidently, CO blobs are a normal feature of cat visual cortex. In the monkey, although the blobs are superficial to the zone of densestgeniculocortical input, the blobs do receive somedirect thalamic input from the intercalated layers of the dLGN (Livingstone and Hubel, 1982; Fitzpatrick et al., 1983; Horton, 1984; Wong-Riley and Carroll, 1984; Lachica and Casagrande, 1992). In the primate visual cortex there is a correlation between the arrangementof geniculocortical inputs and denseCO staining: it is precisely thoseregionsthat receive direct thalamicinput (the blobs, layers IVA, IVC, and VI) where CO stainingis most dense.The radial relationship between geniculocortical inputs and the pattern of CO staining in cat visual cortex appearssimilar to that describedfor the primate and is consistentwith the possibility that blobs may receive some direct thalamic input. Geniculocortical input to the lower tier of layers II-III in the cat is from the C laminae of the dLGN and injection of an anterogradetracer into the C laminae,suchthat eacheye’s laminaeare labeled, results in patchy labeling in the visual cortex with an interpatch spacingof approximately 500 pm (LeVay and Gilbert, 1976; Leventhal, 1979). Thus, the radial and tangential pattern of CO staining in cat visual cortex is consistentwith the notion that denseCO staining may be related to direct input from the dLGN. In addition, the CO blobl in cat visual cortex are most distinct in the region representingthe areacentralis, wherephysiological investigations have found the highest proportion of monocularly driven cells (Albus, 1975), and anatomicalstudies have demonstratedthe most distinct pattern of ocular dominance columns(Anderson et al., 1988). Methodological considerations Since the introduction of techniquesfor flattening the cortical mantleto examinepatternsarrayed tangentialto the cortical laminae (Woolsey and Van der Loos, 1970; Welker and Woolsey, 1974), the use of this type of preparation has grown steadily. When applied to the study of gyrencephalic brains, however, The Journal there is the potential that less than optimal unfolding and flattening strategies may introduce physical distortions or possibly even artifactual histological results. The procedure used in this study for unfolding and flattening the visual cortex, developed by Olavarria and Van Sluyters (1985); produces a well-fixed sheet of cortical gray matter that maintains its cellular integrity (see Fig. 5 in Olavarria and Van Sluyters, 1985) and is resilient to the flattening process. As Olavarria and Van Sluyters indicated, once the white matter is removed and the sulci are opened, the shape of area 17 is similar to the hull of a ship, and this significant intrinsic curvature must be relieved prior to flattening to prevent distortions of area 17. Following the lessons learned by cartographers, who sought to render a distortion-free flat map by introducing a series of discontinuities that relieved the intrinsic curvature of the globe, the visual cortex is flattened by introducing a medial and a lateral cut in the tissue (see Fig. 1B in Olavarria and Van Sluyters, 1985). Although this procedure does result in a discontinuity within area 17, without cuts sufficient to relieve the intrinsic curvature, the process of flattening will result in one of two undesired outcomes. In the case where the visual cortex is adequately fixed, during flattening it will tear at the point of maximum curvature (evidence of this can been seen in Fig. 1A). If the cortex is unfixed, or only very lightly fixed, so that it remains extremely pliable, then insteadof tearing, it may relieve its intrinsic curvature by undergoing differential stretching, compression,and shearing,which must result in distortion of the tissue.Since the cortical mantle is a three-dimensional slab of tissue,this distortion could be very complex. The potential for nonoptimal flattening to distort the tissue has obvious implications for quantitative analysesof tangentially arrayed patterns. Another seriousconcern is the possibility that forcing a piece of unfixed or lightly fixed tissueto lie flat may disrupt its laminar and/or cellular integrity, thereby adversely affecting subsequenthistological analyses.Given these obvious risks, we chose a procedure where the cortical gray matter is firm and cuts are made in the cortical tissueto relieve the intrinsic curvature of this region, rather than a “flat-mount” technique on lightly or unfixed tissue that does not introduce cuts into area 17. In light of the above discussion,a potential, albeit unlikely, concern is that this demonstrationof blobs in cat visual cortex could somehowbe an artifact of the unfolding and flattening procedure. This seemshighly improbable for a number of reasons.First, blobs were found only in the supragranularlayers in area 17-the sameareaand layers where they are found in primates.In addition, the laminar location and 2D pattern of blobs was similar for all casesexamined (see Fig. 2). Finally, when we have examinedthe pattern of CO staining in macaquevisual cortex using the sameperfusion, unfolding, flattening, and COstaining techniquesdescribedin this study (K. Murphy, unpublished observations),the blob pattern, the thick and thin stripes, and the laminar variations in CO staining are the sameas those previously reported (e.g., Horton, 1984). Thus, the logical conclusion from this evidence is that CO blobs are a robust feature of the cat visual cortex. Relationship between blobs and ocular dominance columns The significanceof CO blobs in primatesfor understandingthe organization of the visual cortex follows from their functional relationshipwith other anatomicaland physiological featuresof Vl. To begin to addressthis issuein the cat visual cortex, we examinedthe relationshipof CO blobs to ocular dominancecol- of Neuroscience, June 1995, 7~76) 4205 umnsand found evidencethat blobs are functionally related with the underlying ocular dominancecolumns.Initial analysisof the position of blobs with respectto the ocular dominancecolumns revealed that there are blobs over each eye’s columns; thus, blobs do not simply reflect one eye’s columns. In addition, the location of some blobs in cat visual cortex is not above the center of a labeled ocular dominancedomain, which raisesthe possibility that they may not be functionally eye specific. This arrangementcontrastswith that in monkey visual cortex, where individual blobs are centeredupon and confined entirely within the boundariesof an underlying ocular dominancecolumn (Fitzpatrick and Diamond, 1980; Hendrickson et al., 1981; Horton, 1984). At first glance, these spatial differences might give the impressionthat, unlike in the macaque,in the cat the distribution of blobs is functionally unrelatedto ocular dominancecolumns. However, there are at least two points to considerin interpreting the presentanalysisof the organization of blobs in the cat. First, the apparentdiameter of the blobs in cat visual cortex in comparison to the width of the labeled ocular dominancecolumns is greater than that relationship in monkeys. Thus, it would not be uncommon for the edgesof a blob to overlap the adjacent ocular dominancedomains,even if the blob was centeredupon an unlabeledregion. Second,in macaquevisual cortex, the geniculostriate afferents are strictly segregated,with virtually no overlap of the inputs from the two eyes (Hubel and Wiesel, 1977; Blasdel and Fitzpatrick, 1984; LeVay et al., 1985). In the cat, on the other hand, there is significant overlap of the inputs from the two eyes within layer IV (Shatz and Stryker, 1978), resulting in much lessareathat is strictly monocularthan is the case in the macaque.The regions labeled in cat visual cortex following an eye injection indicate both the zones where only the injected eye hasinputs, as well asthe regionswhere it overlaps with the fellow eye. Only the unlabeled regions reflect the truly monocularaspectsof ocular dominancecolumnsin the cat. Consequently,a simple analysisof the position of blobs above the ocular dominance pattern may fail to reveal if there is a functional link betweenthesetwo features.To examine the functional relationship between blobs and ocular dominancein the cat, in the samemanneras hasbeen done for the monkey, it is necessaryto identify and manipulatethe strictly monocular aspects of the ocular dominancecolumns, then to comparethose monocularity mapswith the array of blobs. To the extent that suchan approachreveals a relationshipbetween blobs and monocular regionsof the ocular dominancepattern, it provides support for the notion that the functional organization of the blobs is similar in cats and monkeys. The question of a functional relationship between blobs and ocular dominancewas addressedin the presentstudy by silencing the activity of one eye, the manipulationusedpreviously to make this functional link in monkeys. The resultsof the present experimentsindicate that CO blobs in cat visual cortex are related to specific aspectsof the ocular dominancecolumns.First, in normally reared cats there are someblobs over the truly monocular regions of layer IV-those areasnot labeledby an eye injection. Second, when the inputs from one eye are silenced, there is a complete lossof blobs over the regions aligned with the monoculardomainsof the enucleatedeye. Thus, the presence of CO blobs in layers II/III of adult cat visual cortex is contingent upon the presenceof active geniculocortical inputs (from either the A, Al, or C laminae), the sameinputs that form the anatomical basis of ocular dominance columns. However, the pattern of CO staining in layers II/III doesnot simply reflect the 4206 Murphy et al. * CO Blobs in Cat Area 17 pattern of active geniculocortical inputs within layer IV, since the regions of dark CO staining in the supragranular layers remain patchy and do not become as extensive as the underlying inputs from the surviving eye. Cortical modularity The demonstration of blobs in the visual cortex of the cat strengthens the evidence for a modular organization within this area. It also raises the intriguing possibility that streams of visual processing, similar to those reported for the primate visual system (Zeki, 1975; Livingstone and Hubel, 1983, 1984a; Shipp and Zeki, 1985; DeYoe and Van Essen, 1985, 1988; Van Essen et al., 1992), may have parallels in the visual systems of these other mammals (Shipp and Grant, 1991; Boyd and Matsubara, 1992). In addition, it is possible that by employing the approach of the present study, blobs may be revealed in the visual cortex of other mammals. Clearly, CO blobs can no longer be referred to as a “primate-Lspecijic” feature of the visual system, and their role as a more general characteristic of mammalian cortical organization must now be considered. The compelling qualitative similarities in the appearance of blobs in cat and macaque visual cortex underscore the idea that the functional organization of this area in these two species may have more in common than was previously proposed. Exactly what the functional significance of the blobs in cat visual cortex is, and how closely the responses of cells within these blobs will match those reported for blob-cells in primate cortex are currently unexplored questions. Blob-cells in primate visual cortex have been reported to exhibit certain characteristic physiological selectivities for color, orientation, and spatial frequency, in addition to their relationship with ocular dominance. The next stage in understanding the significance of blobs in these other mammals will involve exploring the physiological characteristics of blob and interblob cells. It is known that cats have relatively poorly developed color vision (e.g., Weinrich and Zrenner, 1983); consequently, it seems likely that blob cells in cat visual cortex may be similar to those in nocturnal primates (DeBruyn et al., 1993) and not exhibit color selectivity similar to that observed in diurnal primate species. On the other hand, physiological and anatomical information about the organization of the orientation map makes it a possible candidate. Orientation selectivity is well developed and its tangential arrangement is known to be periodic (Schoppman and Stryker, 1981; Singer, 1981; Bonhoeffer and Grinvald, 1991, 1993). Furthermore, the patchy intracortical connections in the supragranular layers of cat cortex (Gilbert and Wiesel, 1979, 1983; Callaway and Katz, 1990; Katz and Callaway, 1992) are known to link like orientations (Gilbert and Wiesel, 1989; T’so et al., 1986). The blobs are interconnected in primate visual cortex (Livingstone and Hubel, 1984b), raising the possibility that the horizontal connections in cat visual cortex also may represent a blob-specific network that links regions with similar orientation selectivity. In addition to studying the tangential pattern of CO staining in cat visual cortex, we, as well as others, have made preliminary investigations of a second species, the ferret (Cresho et al., 1992; Murphy and Stryker, unpublished observations). Sections through surpagranular layers of unfolded and flattened ferret visual cortex revealed patches of dark CO staining regularly spaced within area 17. These patches may relate to features of the ferret visual cortex that are arrayed in a periodic fashion (Redies et al., 1990). The dark patches of CO staining in the ferret were qualitatively similar to those observed in cat visual cortex, although the contrast between blob and interblob regions appeared somewhat less in the ferret. Relative to the tangential area of visual cortex the blobs appear large, so that in comparison with the number of blobs in cat or macaque, there are fewer blobs in ferret visual cortex. This extends the observation made previously by Horton (1984) for primates. The size and number of blobs appears to vary considerably across species. Recent controversy over the degree of segregation of physiological properties into blob and interblob zones in primate cortex (see Merigan and Maunsell, 1993) has raised concerns that the importance of the blobs may not be as great as originally proposed. However, a recent hypothesis about the functional significance of CO blobs that focuses upon the style of computations possible with metabolically more active neurons provides a different perspective, and a possible species-independent function for blob-cells that goes beyond limits set by visual niche differences (Allman and Zucker, 1990). After all, the patchy CO staining observed in supragranular layers of the mammalian visual cortex is an indication that the metabolic activity across its tangential extent is not uniform. How metabolic activity becomes organized in this periodic fashion has remained a largely unstudied question. One thing that is clear is that the development of CO blobs in primate visual cortex is not strictly dependent upon retinal activity, since blobs are present even after removal of both retinae very early in gestation (Kuljis and Rakic, 1990). Furthermore, blobs are present even when the generation of layer IV neurons is disrupted resulting in a failure of geniculocortical afferents to segregate into ocular dominance columns (Kind et al., 1993). Perhaps it is wise to begin to view the significance of the blobs for understanding the organization of the visual cortex, and ultimately vision, in a manner that is consistent with these results-the patchy organization of CO blobs appears to reflect an intrinsic periodicity of the mammalian visual cortex that is independent of certain developmental processes that rely upon retinally based activity (Jones et al., 1991). Our demonstration of CO blobs in cat visual cortex indicates that they are a more ubiquitous feature of mammalian cortical organization than was previously believed and provides a new model system for comparative investigations of their significance for visual cortical organization. At this stage, the regular, patchy nature of CO blobs in the visual cortex of the cat reaffirms the general notion of modularity within the mammalian visual cortex and underscores the importance of species-independent models of functional organization within the visual system . References Albus K (1975) Predominance of monocularly driven cells in the projection area of the central visual field in cat’s striate cortex. Brain Res 89:341-347. Alhan J, Zucker S (1990) Cytochrome oxidase and functional coding in primate striate cortex a hypothesis. Cold Spring Harbor Symp Quant Biol 55:979-982. Anderson PA, Olavarria J, Van Sluyters RC (1988) The overall pattern of ocular dominance bands in cat visual cortex. J Neurosci 8:21833 2200. Beaver C, Murphy KM, Mitchell DE (1992) Patchy expression of c-fos in kitten visual cortex. Sot Neurosci Abstr, 18: 1307. Bonhoeffer T, Grinvald A (1991) Iso-orientation domains in cat visual cortex are arranged in pinwheel-like patterns. Nature 353:429-431. Bonhoeffer T. Grinvald A (1993) The lavout of iso-orientation domains in area 18’of cat visual ‘cortex: optical imaging reveals a pinwheellike organization. .I Neurosci 13:4157-4180. Boyd JD, Matsubara JA (1992) Segregated processing streams in cat visual cortex? Relationship of patchy connectivity to an extrastriate The Journal area, cytochrome-oxidase staining, and local connections. Sot Neurosci Abstr 22:298. Callaway EM, Katz .LC (1990) Emergence and refinement of clustered horizontal connection in cat striate cortex. J Neurosci IO: 1 134-l 153. Cresho HS, Rasco LM, Rose GH, Condo GJ (1992) Blob-like pattern of cytochrome oxidase staining in ferret visual cortex. Sot Neurosci Abstr 18298. DeBruyn EJ, Casagrande VA, Beck PD, Bonds AB (1993) Visual resolution and sensitivity of single cells in the primary visual cortex (Vl) of a nocturnal primate (bush baby): correlations with cortical layers and cytochrome oxidase patterns: J Neurophysiol 69:3-18. DeYoe EA. Van Essen DC (1985) , Segregation of efferent connections -and receptive field properties in visual area V2 of the macaque. Nature 31758-61. DeYoe EA, Van Essen DC (1988) Concurrent processing streams in monkey visual cortex. Trends Neurosci 5:219-226. Dyck R, Beaulieu C, Cynader M (1993) Histochemical localization of synaptic zinc in the developing cat visual cortex. J Comp Neurol 329:53-67. Dyck RH, Cynader MS (1993a) Autoradiographic localization of serotonin receptor subtypes in cat visual cortex: transient regional, laminar, and columnar distributions during postnatal development. J Neurosci 13:4316-4338. Dyck RH, Cynader MS (1993b) An interdigitated columnar mosaic of cytochrome oxidase, zinc, and neurotransmitter-related molecules in cat and monkey visual cortex. Proc Nat1 Acad Sci USA 90:90669069. Fitzpatrick D, Diamond IT (1980) Distribution of acetylcholinesterase in the geniculostriate system of Galago senegalensis and Aotus trivirgatus: evidence for the origin of the reaction product in the lateral geniculate body. J Comp Nemo1 194:703-719. Fitzpatrick D, Itoh K, Diamond IT (1983) The laminar organization of the lateral geniculate body and the striate cortex in the squirrel monkey (Saimiri sciureus). J Neurosci 3:673-702. Gibson AR, Hansma DI, Houk JC, Robinson FR (1984) A sensitive low artifact TMB procedure for the demonstration of WGA-HRP in the CNS. Brain Res 298:235-241. Gilbert CD, Wiesel TN (1979) Morphology and intracortical projections of functionally characterised neurones in the cat visual cortex. Nature 280: 120-l 25. Gilbert CD, Wiesel TN (1983) Clustered intrinsic connections in cat visual cortex. J Neurosci 3: 1116-l 133. Hendrickson AE, Hunt SP, Wu JY (1981) Immunocytochemical localization of glutamic acid decarboxylase in monkey striate cortex. Nature 292:605-607. Horton JC (1984) Cytochrome oxidase patches: a new cytoarchitectonic feature of monkey visual cortex. Philos Trans R Sot Lond [Biol] 304:199-253. Horton JC, Hubel DH (1981) Regular patchy distribution of cytochrome oxidase staining in primary visual cortex of macaque monkey. Nature 2921162-764. Hubel DH, Livingstone MS (1987) Segregation of form, color, and stereopsis in primate area 18. J Neurosci 7:3378-3415. Hubel DH, Wiesel TN (1974) Uniformity of monkey striate cortex: a parallel relationship between field size, scatter, and magnification factor. J Comp Neurol 158:295-305. Jones DG, Van Sluyters RC, Murphy KM (1991) A computational model for the overall pattern of cortical ocular dominance. J Neurosci 11:3794-3807. Kageyama GH, Wong-Riley M (1986) Differential effect of visual deprivation on cytochrome oxidase levels in major cell classes of the cat LGN. J Comp Neurol 246:212-237. Katz LC, Callaway EM (1992) Development of local circuits in mammalian visual cortex. Annu Rev Neurosci 15:31-56. Kind P, Murphy KM, Sengpiel E Blakemore C (1993) Disruption of generation of layer IV in cat area 17 retards geniculate axon segregation but not the emergence of cytochrome-oxidase blobs. Sot Neurosci Abstr 19: 176. Kuljis RO, Rakic P (1990) Hypercolumns in primate visual cortex can develop in the absence of cues from photoreceptors. Proc Nat1 Acad Sci USA 87:5303-5306. Lachica EA, Casagrande VA (1992) Direct W-like geniculate projection to the cytochrome oxidase (CO) blobs in primate visual cortexaxon morphology. J Comp Neurol 319:141-158. of Neuroscience, June 1995, 15(6) 4207 LeVay S, Gilbert CD (1976) Laminar patterns of geniculocortical projection in the cat. Brain Res 113:1-19. Leventhal AG (1979) Evidence that the different classes of relay cells of the cat’s lateral geniculate nucleus terminate in different layers of the striate cortex. Exn Brain Res 37:349-372. Livingstone MS, Hubei DH (1982) Thalamic inputs to cytochrome oxidase-rich regions in monkey visual cortex. Proc Nat1 Acad Sci USA 79:6098-6101. Livingstone MS, Hubel DH (1983) Specificity of cortico-cortical connections in monkey visual system. Nature 304:531-534. Livingstone MS, Hubel DH (1984a) Anatomy and physiology of a color system in the primate visual cortex. J Neurosci 4:309-356. Livingstone MS, Hubel DH (1984b) Specificity of intrinsic connections in primate primary visual cortex. J Neurosci 4:2830-2835. Livingstone MS, Hubel DH (1987a) Connections between layer 4B of area 17 and the thick cytochrome oxidase stripes of area 18 in the squirrel monkey. J Neurosci 7:3371-3377. Livingstone MS, Hubel DH (1987b) Psychophysical evidence for separate channels for the perception of form, color, movement, and depth. J Neurosci 7:3416-3468. Mareschal I, Jones DG, Murphy KM (1994) Tangential distribution of pyramidal neurons in cat areas 17 and 18. Invest Ophthalmol Vis Sci [Suppl] 35:1971. Merigan WH, Maunsell JH (1993) How parallel are the primate visual pathways? Annu Rev Neurosci 16:369-402. Murphy KM, Jones DG, Van Sluyters RC (1989) Analysis of the overall pattern of ocular dominance in normal and visually deprived cats. Invest Ophthalmol Vis Sci [Suppl] 30:30. Murphy KM, Van Sluyters RC, Jones DG (1990) Cytochrome-oxidase activity in cat visual cortex: is it periodic? Sot Neurosci Abstr 16: 292. Murphy KM, Van Sluyters RC, Jones DG (1991 a) Cytochrome-oxidase blobs in cat visual cortex. Invest Ophthalmol Vis Sci [Suppl] 32: 1116. Murphy KM, Van Sluyters RC, Jones DG (1991 b) Analysis of the tangential arrangement of cytochrome-oxidase blobs in cat visual cortex. Sot Neurosci Abstr 17:1088. Price DJ (1985) Patterns of cytochrome oxidase activity in areas 17, 18 and 19 of the visual cortex of cats and kittens. Exp Brain Res 58: 125-133. Olavarria J, Van Sluyters RC (1985) Unfolding and flattening the cortex of gyrencephalic brains. J Neurosci Methods 15: 191-202. Redies C, Diksic M, Rim1 H (1990) Functional organization in the ferret visual cortex: a double label 2-deoxyglucose study. J Neurosci 10:2791-803. Schoen SW, Leutenecker B, Kreutzberg GW, Singer W (1990) Ocular dominance plasticity and developmental changes of 5’-nucleotidase distributions in the kitten visual cortex. J Comp Neurol 69:3-18. Schoppmann A, Stryker MP (1981) Physiological evidence that the 2-deoxyglucose method reveals orientation columns in cat visual cortex. Nature 293:57&576. Shipp S, Grant S (1991) Organization of reciprocal connections between area 17 and the lateral suprasylvian area of cat visual cortex. Vis Neurosci 6:339-355. Shipp S, Zeki S (1985) Segregation of pathways leading from area V2 to areas V4 and V5 of macaque monkey visual cortex. Nature 315: 322-325. Singer W (198 1) Topographic organization of orientation columns in the cat visual cortex. A deoxyglucose study. Exp Brain Res 44:43 l436. Ts’o DY, Gilbert CD, Wiesel TN (1986) Relationships between horizontal interactions and functional architecture in cat striate cortex as revealed by cross-correlation analysis. J Neurosci 6: 1160-l 170. Van Essen DC, Anderson CH, Felleman DJ (1992) Information processing in the primate visual system: an integrated systems perspective. Science 255:419-423. Welker C, Woolsey TA (1974) Structure of layer IV in the somatosensory neocortex of the rat: description and comparison with the mouse. J Comp Neurol 158:437+53. Wong-Riley M (1979) Changes in the visual system of monocularly sutured or enucleated cat demonstrable with cytochrome oxidase histochemistry. Brain Res 171:l l-28. Wong-Riley M, Carroll EW (1984) Effect of impulse blockage on cytochrome oxidase activity in monkey visual system. Nature 307:262264. 4208 Murphy et al. - CO Blobs in Cat Area 17 Wong-Riley M, Riley DA (1983) The effect of impulse blockage on cytochrome oxidase activity in the cat visual system. Brain Res 261: 185-193. Woolsey TA, Van der Loos H (1970) The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex. The description of a cortical field composed of discrete cytoarchitectonic units. Brain Res 17:205-242. Zeki SM (1976) The functional organization of projections from striate to prestriate visual cortex in the rhesus monkey. Cold Spring Harbor Symp Quantitat Biol 40:591-600.