Structure of layer IV in the somatosensory neocortex of the rat: Description and comparison with the mouse

The cytoarchitecture of layer IV of SmI neocortex of the rat has been studied in sections parallel and perpendicular to layer IV. The neurons of layer IV are arranged into discrete multicellular units, called barrels, which we have previously described in both mouse and rat. Using the barrels as markers, it is possible to outline precisely a cytoarchitectonic field—the barrel field—which (1) is confined to SmI. (2) has a consistent appearance from hemisphere to hemisphere, and (3) contains at least 220 barrels. A special region of the barrel field has been identified—the posteromedial barrel subfield (PMBSF)—in which the barrels are arranged in five rows and are larger than elsewhere. However, in contrast to other barrels in the rat barrel field and to the barrels that make up the PMBSF in the mouse, the PMBSF barrels in the rat are filled with small neurons nearly throughout the thickness of layer IV. From experimental evidence, it has been possible for the first time to establish consistent homologies between small groups of neocortical neurons in layer IV. The most surprising finding is that homologous barrels in the PMBSF of two closely related species, the rat and the mouse, have such a different cytoarchitectonic appearance. On the basis of this observation, we suggest that a careful study of these PMBSF barrels in the mouse, the rat and other species may provide a clue to understanding how neocortical neuronal circuits of varying complexity are assembled.     

Cytochrome oxidase staining in the rat SmI barrel cortex

Patterns of cytochrome oxidase (CO) activity were examined histochemically in the rat SmI cortex. Discrete regions of high enzymatic activity were centered upon the granule cell aggregates (barrels) in layer IV. Those barrels which correspond to the mystacial vibrissae and make up the posteromedial barrel subfield (PMBSF) were especially interesting in that CO staining revealed distinct metabolic subdivisions which do not have an easily demonstrable cytoarchitectonic counterpart. By analogy with the barrels in mouse PMBSF and with the cytoarchitectonically distinct barrels representing the smaller sinus hairs in the rat we propose that regions of high CO activity denote the "hollow" of the rat PMBSF barrels. In accord with previous physiological studies demonstrating a vertical organization in the rodent barrel cortex, we also noted columns of intense CO activity extending from layer VI through sublamina Vb. The centers of these columns coincided with the centers of the barrels in layer IV. In tangential sections through the infragranular laminae the segmentation of CO-positive zones was less distinct than in layer IV and appeared as bands of heightened activity oriented like the five rows of layer IV barrels. Highly reactive somata and dendrites were observed in both the granular and infragranular CO barrels indicating that some of the increased activity of these regions reflects oxidative metabolism of cortical neurons per se. These patterns of metabolic activity underscore the vertical and horizontal organization of the SmI vibrissa cortex and suggest that neurons located within the central core of a column have functional properties distinct from those located in zones where individual columns interface.     

Autoradiographic localization of gamma-aminobutyric acid receptors in mouse barrel field

The distribution of high-affinity gamma-aminobutyric acid (GABA) receptors in mouse posteromedial barrel subfield (PMBSF) in layer IV of the somatosensory (SI) cortex was studied using [3H]muscimol autoradiography in vitro. A qualitative study revealed a heterogeneous distribution in the density of [3H]muscimol binding in the barrel field. In the barrel sides and septum between the barrels [3H]muscimol binding exhibits the lowest level of labelling in the entire tissue. In comparison, [3H]muscimol binding in the hollows is considerably higher than in the barrel sides although the level is not homogeneous. These findings suggest that the barrel hollows are much richer in GABA receptors than the surrounding barrel sides and septa.     

Spatiotemporal distribution of proteoglycans in the developing rat's barrel field and the effects of early deafferentation

The isolectin Vicia villosa B(4) (VV) selectively recognizes N-acetyl-galactosamine-terminal glycoconjugates that form perineuronal nets (PNNs) around a subset of neurons in the cerebral cortex. PNNs are thought to participate in the guidance of incoming thalamic axons and in the posterior stabilization and maintenance of synaptic contacts. Here we examine the spatial and temporal distribution of biotinylated VV in tangential sections through layer IV of the posteromedial barrel subfield in the primary somatosensory cortex (PMBSF) of rats ranging from postnatal day (P)3 to P60, which underwent unilateral deafferentation of whiskers at birth. In the afferented hemisphere, labeling first appears at P5, with a diffuse distribution, probably associated with neuropil, inside PMBSF barrels. VV distribution remains diffuse during the following week, and declines around P17. From P24 onward, however, proteoglycans form PNNs around cell bodies preferentially localized in septal regions of the PMBSF. In the contralateral, deafferented PMBSF the diffuse labeling also appears on P5, but first develops into elongated, homogeneous stripes, which disappear after P24, leaving only scattered cell bodies along layer IV. Our results indicate that proteoglycans appear simultaneous to barrel formation in the developing rat while segregation of PNNs to septal cells might be driven by afferent activity.     

Ultrastructure and synaptic contacts in barrels of mouse SI cortex.

Barrels in the posteromedial barrel subfield (PMBSF) in layer IV of mouse somatosensory cortex are consistently identifiable and somatotopically related to the large whiskers on the animal's snout32,38. This is a study of the ultrastructure - in particular the synaptic connections - of mouse PMBSF barrels. A technique was developed which enabled the precise selection for electron microscopy of specific regions of barrels identified in the light microscope. Barrel "sides" containing a high density of neuronal cell bodies and myelinated axons, contrasted with barrel "hollows" in which these elements were relatively sparse. The types and frequencies of synapses were examined in series of thin sections through sides and/or hollows of barrels B2, C3, C7, D4, E4, E5, and E8 from 6 mice. In all, 3042 synapses from 20 fields, each at least 10 mum x 10 mum x 4.8 mum in size, were classified. Four distinct kinds of presynaptic terminal were identified: (1) darkly, and (2) lightly stained, asymmetrically synapsing terminals (AD, AL respectively), and (3) darkly, and (4) lightly stained, symmetrically synapsing terminals (SD, SL respectively). Synapses of these terminals were distributed in roughly similar proportions throughout the neuropil of barrel sides and hollows. Thus in all barrel regions approximately 92% of the synapses in neuropil were made by AD terminals, 1.0% by AL, 2.4% by SD and 4.0% by SL terminals.     

Development of the barrels and barrel field in the somatosensory cortex of the mouse

Barrels of the PMBSF of the mouse somatosensory cortex become apparent in Nissl-stained tangential sections simultaneously, on the fourth postnatal day. At this time they are miniatures of those in the adult and are situated in the deepest sublamina of the trilaminar cortical plate. An early barrel appears as a patch of decreased cell density: the prospective hollow of the barrel. Septa become noticeable during the sixth postnatal day. From that period to adulthood, the relative contribution of the PMBSF to the total cortical surface area increases — an increase that goes against one's expectation: the barrel related periphery matures very early and so does the central, lateral region of the cortex. Barrel growth parallel to the pial surface is greater along the major axes than along the minor axes. By using the barrels to identify prospective layer IV in immature cortex, we could determine that layers V and VI attain their adult height during the sixth postnatal day — an age when prospective layers I-IV are only half their adult height. The onset of barrel formation coincides with the moment after which injury to the pertinent somatosensory periphery (the vibrissal papillae) no longer causes profound alterations in barrel morphology.     

Acute whisker removal reduces neuronal activity in barrels of mouse SmL cortex

The autoradiographic 2-deoxy-D glucose (2-DG) method has been used to map relative changes in metabolic activity in the CNS during various functional states (Plum et al., '76). Here we describe the application of the 2-DG method to assay regional activity in the posteromedial barrel subfield (PMBSF) region of the mouse SmI cortex after acute removal of mystacial vibrissae. One day prior to isotope injection, various combinations of vibrissae (e.g., all vibrissae, row-C only, rows-B and -D only) were plucked from adult male Swiss Webster mice under anesthetic. The next day, 5 muCi of 14C-2-DG were injected into a tail vein, and the mice were allowed to actively explore an empty cage for 45 minutes. The animals were then sacrificed, the brains quickly removed, frozen, and sectioned either parallel or perpendicular to the pia at 80 mum in a cryostat. The sections were mounted, dried on coverslips, and were used to expose X-ray film, after which the sections were stained with thionin and the X-ray film developed. The tissue sections and matching autoradiograms were compared directly from photomicrographs of each. The autoradiograms showed areas of higher activity in barrels for which corresponding vibrissae were present and lower activity in barrels for which appropriate vibrissae were missing. In tangential sections from animals with all vibrissae intact, the PMBSF was uniformly and consistently higher in activity than in cases with all vibrissae missing. The removal of row-C or rows-B and -D resulted in strips of decreased activity in the corresponding PMBSF rows. The same patterns of increased or decreased activity were also seen in sections normal to the pia, but the changes in activity, while greatest in layer IV, extended through all layers of the cortex. Finally, in a number of the autoradiograms, density patterns could be recognized which later were shown to relate directly to sides of individual barrels. The results indicate: (1) Acute removal of the peripheral vibrissal hairs is sufficient to deprive the related contralateral cortical barrel neurons of normal activity. Thus in the mouse somatosensory system it may be possible to determine the relative importance of sensory deprivation and neonatal peripheral lesions in normal cortical development. (2) The barrels are part of a functional cortical columnar organization similar to that in other sensory systems. And, (3) the 14C-2-DG-X-ray technique is sufficiently sensitive to reveal parts of individual barrels in autoradiograms and thus, with some modification, may be suitable for the study of small populations of neurons.     

Differential effects of methylmercury intoxication in the rat's barrel field as evidenced by NADPH diaphorase histochemistry

In the present study, we investigated the effects of mercury intoxication on the structure of the posteromedial barrel subfield (PMBSF) in the primary somatosensory cortex (SI) of adult rats, as revealed by histochemical reactivity to the enzyme NADPH diaphorase (NADPH-d). Enzymatic reactivity in the neuropil inside barrels was drastically reduced in intoxicated animals, suggesting that the synthesis and/or transport of the nitric oxide synthase enzyme can be altered in acute mercury intoxication. However, the cell bodies and dendrites of barrel neurons, also strongly reactive to the enzyme, were spared from the mercury's deleterious effects.     

Cortical modules in the posteromedial barrel subfield (Sml) of the mouse

An antibody to MAP2 was used on sections through the posteromedial barrel subfield (PMBSF) of primary somatosensory cortex to reveal the distributions of cell bodies and dendrites. It was found that apical dendrites of layer VI neurons form irregular bundles or sheets that break up in layer IV, where most of these dendrites form their terminal tufts. In contrast, the apical dendrites of layer V pyramidal neurons form clusters that ascend into layer II/III where they are joined by apical dendrites of the superficial pyramidal neurons. In layer IV the clusters of the layer V apical dendrites are more concentrated in barrel walls than in hollows. Thus, in layer IV the average center to center spacing between the clusters is about 25 m?m in the barrel hollows, and about 22 m?m in the barrel walls. In part, this differential distribution of the apical dendritic clusters is brought about because the apical dendrites of layer V pyramids beneath the periphery of the barrel hollows arc towards the barrel walls as they pass from layer V into layer IV. Based on previous analyses of the three-dimensional organization of the primary visual areas in the monkey, cat, and rat, it has been proposed that neurons in these cortices are organized into modules that are centered on the clusters of apical dendrites belonging to layer V pyramidal neurons. Mouse PMBSF cortex is composed of similar pyramidal cell modules and the organization of neurons in these modules is similar to that in visual cortex. This suggests that the pyramidal cell modules are fundamental neuronal units that exist throughout the cerebral cortex, and implies that the various functional areas of the cortex in different species are organized according to a common, basic plan. © 1993 Wiley-Liss, Inc.     

Transient patterns of GAP-43 expression during the formation of barrels in the rat somatosensory cortex

The development of the rat barrel field cortex was investigated with an antibody to the axonal membrane-specific phosphoprotein GAP-43 in order to examine the developmental pattern of afferent projections, and with cytochrome oxidase histochemistry and Nissl stains to reveal the morphogenesis of cortical barrels. On the first two days after birth, GAP-43 immunostaining in the cortical plate was light and diffuse, then became intense in the presumptive layer IV of the parietal cortex on PND3 (day of birth = PND0). Immunoreactive densities were visible as small, focal patches within the centers of prospective barrels. These densities increased in size and intensity over the next few days and then diminished abruptly. On PND7, the distribution of GAP-43 was coextensive with barrels, as defined by cytochrome oxidase histochemistry and Nissl staining. GAP-43 virtually disappeared from the barrels after PND7. From the second postnatal week, GAP-43 immunostaining was evident in the septa between barrels and in the dysgranular regions of SI cortex. This pattern of GAP-43 distribution was complementary to the pattern of cytochrome oxidase activity, and persisted into maturity. In an attempt to identify possible source(s) of GAP-43 positive afferents in the developing barrels, we examined the effects of altering the sensory periphery on the distribution of GAP-43 immunostaining in the cortex. Rat pups had row C whiskers cauterized on PND0 and were sacrificed on PND3 or PND5. Whereas immunopositive densities corresponding to intact whiskers developed in a normal, punctate pattern, cortical representation of the lesioned whiskers formed a continuous band of labeling that was evident as early as PND3. We argue that the disjunctive expression of GAP-43 in the barrel field reflects the pattern of distribution of afferents (most likely from the ventro-basal thalamic nucleus) to the barrel field cortex, and that this pattern may be instructive in the formation of barrels as cytoarchitectonic units. The rapid alteration in patterns of immunostaining following whisker lesions lends further support to the conclusion that the "barrel template" is conveyed to the neocortex by incoming afferents. The possible significance of the transient expression of GAP-43 in the maturing barrel field is discussed.     

Differential origin of projections from SI barrel cortex to the whisker representations in SII and MI

We have previously shown that projections from SI barrel cortex to the MI whisker representation originate primarily from columns of neurons that are aligned with the layer IV septa. SI barrel cortex also projects to SII cortex, but the origin of these projections has not been characterized with respect to the barrel and septal compartments. To address this issue, we injected retrograde tracers into the SII whisker representation and then reconstructed the location of the labeled neurons in SI with respect to the layer IV barrels. In some animals, two different tracers were injected into the whisker representations of SII and MI to detect double-labeled neurons that would indicate that some SI neurons project to both of these cortical areas. We found that the projections to SII cortex originate from sites that are uniformly distributed throughout the extragranular layers of barrel cortex. In cases in which different tracers were injected in SII and MI, double-labeled neurons appeared above and below the layer IV septal compartment and at sites aligned with the boundaries of the layer IV barrels. To the extent that the columns of neurons aligned with the barrel and septal compartments represent functionally distinct circuits, these results indicate that SII receives information from both circuits, whereas MI receives inputs primarily from the septal circuits.     

Effects of cholinergic depletion on neural activity in different laminae of the rat barrel cortex

The purpose of these experiments was to determine the effects of cholinergic depletion on spontaneous and evoked activity of neurons in the different layers of the posteromedial barrel subfield (PMBSF) of the rat somatosensory cortex. Acetylcholine neurons in nucleus basalis of Meynert (NBM) were selectively lesioned with an immunotoxin (IT), 192 IgG-saporin. Spontaneous activity was significantly lower in layers II-III, Va, and VI in IT-injected animals compared to control animals. Evoked activity was significantly lower in layers II-III, IV, Vb, and VI of IT-injected animals compared to control animals. The largest difference was observed in layer Vb. Thus, cholinergic depletion causes significant changes in the magnitude of spontaneous and evoked activity but these differences are not completely in register with one another.     

Septal columns in rodent barrel cortex: functional circuits for modulating whisking behavior

In rodents, each mystacial whisker is represented in the granular layer of primary somatosensory (SI) cortex by a compact cluster of cells known as a barrel, and barrels are separated from each other by domains that are called septa. Vertical columns of neurons aligned with each barrel act as a functional assembly to process information from a "principal" whisker, but a functional role has not been identified for vertical columns of neurons that are aligned with the septa. To determine whether these septal columns provide the main source of projections to primary motor (MI) cortex, we placed retrograde tracers in MI cortex and analyzed the location of the retrogradely labeled neurons with respect to the septal and barrel compartments of SI barrel cortex. In cases in which SI barrel cortex was sectioned tangentially, retrogradely labeled neurons in the extragranular layers of SI were plotted and superimposed onto reconstructions of the layer IV barrel field. In each of these cases, most labeled neurons were located above or below the septal regions of layer IV. When SI barrel cortex was sectioned coronally, we observed multiple columns of labeled SI neurons that were vertically aligned with the septal zones of layer IV. These results indicate that columns of neurons that are vertically aligned with the septa, or septal columns, are functionally linked by virtue of their projections to MI cortex. We hypothesize that these septal columns represent an interconnected and functionally distinct circuit that transmits information to MI and other brain regions involved in motor control.     

Lectins demarcate the barrel subfield in the somatosensory cortex of the early postnatal mouse

Plant lectins were used to examine the disposition of glycosylated molecules in vibratome sections through the barrel subfield of mouse somatosensory cortex at selected times during postnatal development. The peroxidase conjugates of peanut agglutinin (PNA, specific for N-acetylgalactosamine), concanavalin A (specific for mannose), and wheat germ agglutinin (specific for N-acetylglucosamine and N-acetylneuraminic acid) were used to study lectin binding in aldehyde-fixed tissue sections of cortex. Following peroxidase cytochemistry and light microscopy, it was found that all three lectins bound in the region of the barrel subfield as early as postnatal day 3 (day of birth = postnatal day 1). The lectins bound to the prospective sides and/or septae of individual barrels in preference to the prospective hollows. This lectin demarcation of the barrel field occurred prior to the detection of this region with cresyl violet staining and was still demonstrable on postnatal day 6, when the individual barrels became discernible with cresyl violet. This suggests that the lectin binding material is present before the barrel field becomes a fully formed and organized region. A decrease in lectin affinity for binding sites in these tissue sections occurs during postnatal development (Cooper and Steindler: Soc. Neurosci. (Abstr.) 10: 43a, '84) and this study demonstrates that lectins do not delineate the barrel field of more mature animals (2-3 months old), whereas barrels can be detected with cresyl violet at this time. A preliminary electron microscope analysis of the postnatal day 6 somatosensory cortex demonstrates that the lectin PNA binds to elements of the forming neuropil and also to Golgi apparatus intermediate saccules in neuronal cells. The prospective barrel field can be detected with lectins during a critical period in development in which alterations can occur in the barrel field in response to peripheral deprivation (Jeanmonod et al: Neuroscience 6:1503-35, '81) and therefore we suggest that the glycans visualized with lectin-peroxidase conjugates denote possible candidates for molecules involved in shaping barrel structure.     

Differential synaptic vesicle protein expression in the barrel field of developing cortex

The distribution of four proteins associated with synaptic vesicles, SV2, synaptophysin, synapsin I, and rab3a, was investigated during postnatal development of the posteromedial barrel subfield (PMBSF) in the rat somatosensory cortex. A distinct progression in the appearance of the different synaptic vesicle proteins within the PMBSF was observed. SV2, synapsin I, and synaptophysin revealed the organization of the barrel field in the neonate. This early demarcation of the cortical representation of the vibrissal array coincides with the earliest known age for the emergence of the cytoarchitectonic organization of this region. In contrast, rab3a did not delimit the barrels until the end of the 1st postnatal week, coincident with the known onset of adult-like physiological activity and the loss of plasticity in afferents to this region. In addition, the appearance of the different synaptic vesicle proteins occurred earlier within the PMBSF than in the adjacent extra-barrel regions of the cortex. These results show that the molecular differentiation of synaptic fields across the cortex is not a homogeneous and synchronous process in terms of synaptic vesicle protein expression. Because these proteins act together in mature synapses to ensure the regulated release of neurotransmitters, our results suggest that this temporo-spatial asynchrony may underlie different potentials for synaptic activity and thus contribute to the development of cortical maps. © 1996 Wiley-Liss, Inc.     

Spatial organization of facial vibrissae and cortical barrels in the guinea pig and golden hamster

The arrangements of vibrissae in guinea pigs and golden hamsters were previously reported to be different from those in mice and rats. Whereas the mystacial pads in mice and rats include four straddlers and five rows of vibrissae, guinea pigs were described to possess six rows of irregularly aligned mystacial vibrissae and no straddlers, and golden hamsters to include seven vibrissal rows and also no straddlers. We found that all of these four species possess similar vibrissal arrangements within the mystacial pad. To demonstrate this similarity, we developed a new method of sinus hair visualization in flattened and cleared preparations of the mystacial pad. Intrinsic muscles of the mystacial pad that were revealed in thick histological preparations showed clearly the structural and functional relationships between straddlers and vibrissal rows. To verify this finding, and to extend the knowledge of vibrissal cortical representations in guinea pigs and golden hamsters, we have investigated the spatial organization and the functional vibrissal representations of barrels in the posteromedial barrel subfield (PMBSF) of these rodents. The barrel morphology was clearly preserved in Nissl-stained sections and sections processed for cytochrome oxidase of flattened cerebral cortices. We demonstrate that the vibrissal arrangement in the mystacial pad is replicated in the PMBSF of guinea pigs and golden hamsters and that this arrangement is similar to that found in mice and rats. To facilitate comparative studies, these findings strongly recommend the use, in guinea pigs and golden hamsters, of the same classifications and nomenclatures that are used in mice and rats to describe mystacial vibrissae and cortical barrels. J. Comp. Neurol. 385:515–527, 1997. © 1997 Wiley-Liss, Inc.     

Functional organization in cortical barrels of normal and vibrissae-damaged mice: a (3H) 2-deoxyglucose study

The large mystacial vibrissae on the faces of rodents have punctate representations in all stations in the central trigeminal pathway, including layer IV of the somatosensory cortex (SmI). The cortical whisker correlates, multicellular units termed barrels, are not present at birth, and damage to the vibrissae during the first postnatal week results in altered adult cytoarchitectonics. The anatomical effects of vibrissae damage in the cortex have been well documented; here, we investigated the functional organization of altered SmI barrels with a high-resolution 2-deoxyglucose (2-DG) technique (Durham et al., '81, J. Neurosci. 1:519). The middle row of vibrissae was cauterized in 1-, 2-, 3-, 4-, or 5-day-old mice, and the animals were allowed to survive to sexual maturity. Various combinations of vibrissae were clipped acutely 24 hours prior to injection of 2-4 mCi of (3H)2-DG. Mice actively explored an empty cage for 60 minutes, stimulating the remaining vibrissae. The mice then were perfused and their brains prepared for paraffin histology and emulsion autoradiography. In tangential sections through layer IV, patterns of neuropil and cell body labeling were analyzed with respect to barrel cytoarchitecture in normal and vibrissae-damaged mice. In both control and experimental animals, patterns of neuropil and cell somata label corresponded exactly to barrel boundaries, whether normal or altered by vibrissae damage. Only those barrels for which vibrissae were intact had high levels of label, with anterior barrels more heavily labeled. Many neurons in the septa between these barrels and the adjacent barrels were labeled also. We found slightly higher neuropil label in the cortical zone corresponding to the damaged zone on the face in animals lesioned at any time. These data indicate that physiological somatotopy in vibrissae-damaged animals matches the anatomical cytoarchitecture.     

Effects of neonatal 6-hydroxydopamine treatment upon morphological organization of the posteromedial barrel subfield in mouse somatosensory cortex

While recent studies indicate that proposed regulation of visual cortex plasticity by norepinephrine (NE) probably involves 6-hydroxydopamine (6-OHDA) effects other than NE depletion, reports of diminished neuronal maturation and reduced postweaning sensitivity to environmental conditions in animals treated with anti-adrenergic compounds continue to suggest a role for NE in neocortical development. To assess its possible role in development of a highly organized neocortical structure, the effects of postnatal 6-OHDA administration upon development of the somatosensory cortical posteromedial barrel subfield (PMBSF), which subserves the large facial whiskers, were observed in mice with whiskers in the middle row of the face removed unilaterally. Treatment with 6-OHDA caused 96-98% depletion of NE in parietal cortex. There were no effects of (or lesion interactions with) 6-OHDA treatment on barrel size in measures of Nissl-stained neurons, and 6-OHDA effects on numerical measures of dendritic branching of Golgi-impregnated non-pyramidal neurons in PMBSF were negligible. There were, however, effects of 6-OHDA treatment upon the highly ordered arrangement of dendrites within barrels. In 6-OHDA-treated animals, the class I (spiny and sparsely spined) cell dendrites were less attracted to the barrel hollow. In controls, for class I cells with their somata within the barrel wall, there was a high correlation between the distance from the soma to the hollow and the amount of dendrite in the wall, reflecting the distance the dendrite traverses to the hollow. In 6-OHDA-treated animals, this correlation was absent, i.e., cells at any distance from the hollow tended to have a high percentage of dendrite in the wall.(ABSTRACT TRUNCATED AT 250 WORDS)     

Comparative anatomical studies of the Sml face cortex with special reference to the occurrence of “barrels” in layer IV

In the SmI cortex of mice and rats there are cytoarchitectonically identificable groups of cells — called barrels — some of which have been shown to be directly related to whiskers and other sensory hairs on the contralateral face. In this study we have used a comparative approach to determine the incidence and variation of the barrels. The brains of 27 mammalian species have been examined histologically to determine whether barrels exist in layer IV of what is known or likely to be the face area of SmI. Thick sections (50–100 μm) were taken tangential to the pia overlying SmI and stained with thionin. The patterns of facial whiskers were also mapped by dissection of the facial skin. Barrels were seen only in brains of species belonging to three of the seven mammalian orders examined. We have confirmed Weller's ('72) observation of barrels in the Australian brush-tailed possum but have not found barrels in two marsupials from the western hemisphere. Barrels were demonstrable in representatives of four of five rodent suborders examined and in the rabbit. From the study of the rodent brains, a number of trends emerge. (1) The organization of the barrel fields is “dictated” by the organization of the sensory periphery. Animals with five rows of large mystacial (moustache-like) vibrissae have five rows of PMBSF (Posterom edial barrel sub-field) barrels. (2) The barrels are confined to layer IV of (what is known or likely to be) the SmI face area. The pattern and cortical location of the barrel field is consistent among different specimens of the same species. (3) Certain behavioral patterns do not preclude the existence of the barrels. Species which possess well developed visual systems and behaviors (e.g., grey squirrel) and forms which do not actively explore the environment by whisking their vibrissae (e.g., guinea pig) have barrels. (4) Within a given rodent suborder, the barrels become more difficult to identify, as the brains become larger. We have not yet been able to demonstrate barrels in the largest rodent, the capybara.     

Gradual changes in the structure of the barrels during maturation of the primary somatosensory cortex in the rat

The maturation of the barrel field in the primary somatosensory cortex was observed in Nissl-stained preparations from rats ranging in age from 12 days to 1.5 years postpartum. Prior to the 20th day, the barrels in the rat resemble those of the mouse and have distinct cell-sparse hollows that are surrounded by cell-dense sides. They span the full thickness of layer IV. Between the 20th and 34th days, the barrels in only the posteromedial part of the barrel field gradually change and the distinction between the hollows and sides is lost throughout all but the deepest part of layer IV. The resulting mature barrels are relatively indistinct and have a uniformly high cell density that extends well into the supragranular layers. In contrast, the barrels in the anterolateral part of the barrel field remain essentially unchanged. The remodeling apparently is not due passively to cortical growth because, by P20, the thicknesses of the cortical layers and the dimensions of the barrels are virtually the same as in the adult. Several mechanisms are considered that may account for the changes. These include a redistribution of the neurons that originally were in barrel sides; a reduction in the neuropil between the neurons that originally were within hollows; and differential growth of layer IV dendrites. The changes in the barrel structure may be related to the differentiation and quantity of innervation in the hairy skin between the vibrissae.