A quantitative study of synaptic patterns in turtle visual cortex

The turtle visual system mediates a complex spatiotemporal transformation of visual stimuli into cortical activity

The three-layered visual cortex of turtle is characterized by extensive intracortical axonal projections and receives non-retinotopic axonal projections from lateral geniculate nucleus. What spatiotemporal transformation of visual stimuli into cortical activity arises from such tangle of malleable cortical inputs and intracortical connections? To address this question, we obtained band-pass filtered extracellular recordings of neural activity in turtle dorsal cortex during visual stimulation of the retina. We discovered important spatial and temporal features of stimulus-modulated cortical local field potential (LFP) recordings. Spatial receptive fields span large areas of the visual field, have an intricate internal structure, and lack directional tuning. The receptive field structure varies across recording sites in a distant-dependent manner. Such composite spatial organization of stimulus-modulated cortical activity is accompanied by an equally multifaceted temporal organization. Cortical visual responses are delayed, persistent, and oscillatory. Further, prior cortical activity contributes globally to adaptation in turtle visual cortex. In conclusion, these results demonstrate convoluted spatiotemporal transformations of visual stimuli into stimulus-modulated cortical activity that, at present, largely evade computational frameworks.

Subpial and stellate cells: two populations of interneurons in turtle visual cortex

Turtle visual cortex has three layers and receives direct input from the dorsolateral geniculate complex of the thalamus. The outer layer 1 contains several populations of interneurons, but their physiological properties have not been characterized. This study used intracellular recording methods followed by filling with Neurobiotin to characterize the morphology and physiology of two populations of layer 1 interneurons. Subpial cells have somata positioned in the outer third of layer 1 and dendrites confined within the band of geniculate afferents that runs from lateral to medial across visual cortex. Their dendrites are composed of a sequence of many beads or varicosities separated by intervaricose segments. They have membrane time constants of tau(o) = 45.5 +/- 5.2 ms and electrotonic lengths of 1.1 +/- 0.2. Subpial cells show spike rate adaptation in response to intracellular current pulses. Stellate cells have somata located in the inner two-thirds of layer 1 and, less frequently, in layers 2 and 3. Their dendrites extend in a stellate configuration across the cortex. They are smooth or sparsely spiny, but never bear distinct varicosities. They have membrane time constants of tau(o) = 155.1 +/- 12 ms and electrotonic lengths of 3.8 +/- 0.5. They show little spike rate adaptation in response to intracellular current pulses. The positions of the two populations of cells in visual cortex and their physiological properties suggest that subpial cells may participate in a feedforward inhibitory pathway to pyramidal cells, whereas stellate cells are involved in feedback inhibition to pyramidal cells.

Neurons of the medial cortex outer plexiform layer of the lizard Podarcis hispanica: Golgi and immunocytochemical studies

The study of Golgi-impregnated lizard brains has revealed a scarce but heterogeneous neuronal population in the outer plexiform layer of the medial cortex. Some of the neuronal types detected here resemble the neurons of the dentate molecular layer of the mammalian hippocampus. According to their morphology, five intrinsic neuronal types have been clearly identified: short axon aspinous bipolar neuron (type 1, or sarmentous neuron), short axon aspinous juxtasomatic neuron (type 2, or coral neuron), short axon sparsely spinous multipolar neuron (type 3, or stellate neuron), short axon sparsely spinous juxtasomatic multipolar neuron (type 4, or deep stellate neuron), and sparsely spinous juxtasomatic horizontal neuron (type 5, or couchant neuron). Most neuronal types were identified as gamma-aminobutyric acid (GABA) and parvalbumin immunoreactive, and are thus probably involved in medial cortex inhibition. Moreover, a small fraction of them displayed beta-endorphin immunoreactivity. The distribution of these neuronal types is not uniform in the laminae of the outer plexiform layer. Type 1 (sarmentous) and type 3 (stellate) neurons overlap the axonal field projection coming from the dorsal cortex and the thalamus, whereas types 4 (deep stellate) and 5 (couchant) neurons overlap ipsi- and contralateral dorsomedial projection fields as well as raphe serotoninergic and opioid immunoreactive axonal plexi. Thus, these neuronal types may be involved in the control of specific inputs to the medial cortex by presumably feed-forward inhibition; nevertheless, feed-back inhibition may also occur regarding type 4 (deep stellate) neurons that extend deep dendrites to the zinc-rich bouton field.

Appearance of putative amino acid neurotransmitters during differentiation of neurons in embryonic turtle cerebral cortex

Pyramidal and nonpyramidal neurons can be recognized early in the development of the cerebral cortex in both reptiles and mammals, and the neurotransmitters likely utilized by these cells, glutamate and gamma-aminobutyric acid, or GABA, have been suggested to play critical developmental roles. Information concerning the timing and topography of neurotransmitter synthesis by specific classes of cortical neurons is important for understanding developmental roles of neurotransmitters and for identifying potential zones of neurotransmitter action in the developing brain. We therefore analyzed the appearance of GABA and glutamate in the cerebral cortex of embryonic turtles using polyclonal antisera raised against GABA and glutamate. Neuronal subtypes become immunoreactive for the putative amino acid neurotransmitters GABA and glutamate early in the embryonic development of turtle cerebral cortex, with nonpyramidal cells immunoreactive for GABA and pyramidal cells immunoreactive for glutamate. The results of controls strongly suggest that the immunocytochemical staining in tissue sections by the GABA and glutamate antisera corresponds to fixed endogenous GABA and glutamate. Horizontally oriented cells in the early marginal zone (stages 15-16) that are GABA-immunoreactive (GABA-IR) resemble nonpyramidal cells in morphology and distribution. GABA-IR neurons exhibit increasingly diverse morphologies and become distributed in all cortical layers as the cortex matures. Glutamate-immunoreactive (Glu-IR) cells dominate the cellular layer throughout development and are also common in the subcellular layer at early stages, a distribution like that of pyramidal neurons and distinct from that of GABA-IR nonpyramidal cells. The early organization of embryonic turtle cortex in reptiles resembles that of embryonic mammalian cortex, and the immunocytochemical results underline several shared as well as distinguishing features. Early GABA-IR nonpyramidal cells flank the developing cortical plate, composed primarily of pyramidal cells, shown here to be Glu-IR. The earliest GABA-IR cells in turtles likely correspond to Cajal-Retzius cells, a ubiquitous and precocious cell type in vertebrate cortex. Glutamate-IR projection neurons in vertebrates may also be related. The distinctly different topographies of GABA and glutamate containing cells in reptiles and mammals indicate that even if the basic amino acid transmitter-containing cell types are conserved in higher vertebrates, the local interactions mediated by these transmitters may differ. The potential role of GABA and glutamate in nonsynaptic interactions early in cortical development is reinforced by the precocious expression of these neurotransmitters in turtles, well before they are required for synaptic transmission.(ABSTRACT TRUNCATED AT 400 WORDS)

Richness of environment affects the number of contacts formed by boutons containing flat vesicles but does not alter the number of these boutons per neuron

A recent quantitative analysis of cat visual cortex has demonstrated that the numerical density (Nv) of symmetrical synaptic contacts formed by boutons containing flat vesicles (FS synapses) is nearly twice as large in animals raised in isolation (impoverished condition: IC) as in animals raised in a colony (enriched condition: EC). Although some FS synapses have been shown to be cholinergic there is evidence that many, indeed the vast majority, are GABAergic. In order to estimate whether the change in the Nv of FS contacts was accompanied by a change in the number of boutons containing GABA, we have incubated sections of tissue from both groups of animals in an antiserum for GAD. In spite of the large increase in the number of FS contacts in impoverished cortex, we saw no obvious change in the apparent amount of labelled GAD terminals. In retrospect we realized that though the amount of labelled GAD terminals might reasonably be expected to reflect the number of F-boutons, it might not correspond so closely to the number of contacts formed by these boutons (which is what we had measured in the previous study): The richness of the environment could conceivably affect the number of contacts formed by the F-boutons without affecting the number of boutons! We thus extended our study by estimating the number of F-boutons in the two conditions. For the total cortical thickness, the Nv of F-boutons is only 17% lower (P less than .05) in enriched than in impoverished cats. The diameter of the boutons is 6% larger (P less than .001) in the enriched cortex. Because the F-boutons become fewer in number as they become larger in size, the total percentage volume occupied by these boutons does not change between the two experimental conditions. We conclude that this is the reason why there appears to be no change in the general amount of GAD label between the two groups of cats. More importantly, since the Nv of neurons is also 17% lower in enriched cortex, the number of F-boutons per neuron (and presumably the total number of F-boutons in the visual area) actually remains unchanged. In contrast, the previous study showed that the number of FS contacts per neuron is significantly decreased in enriched cortex. It follows that the number of contacts formed by each bouton must be altered.(ABSTRACT TRUNCATED AT 400 WORDS)

Development of GABA responsiveness in embryonic turtle cortical neurons

The whole-cell patch-clamp method was used to study the development of functional GABA receptors in cortical neurons dissociated from embryonic turtles. GABA elicited an increase in membrane conductance, even from cells obtained from the earliest stages of corticogenesis. The GABA-mediated conductance had a mean value 7.4 times greater than membrane 'leak' conductance and increased with developmental age. In all stages studied, the response inverted polarity at a value approximating ECl- and was blocked by applications of bicuculline, suggesting that it was mediated by GABAA receptors. GABA receptors are thus present and functional very early in corticogenesis, preceding electrogenesis, synaptogenesis, and full neuronal differentiation.

Organization of glutamate-like immunoreactivity in the rat superficial dorsal horn: light and electron microscopic observations

Glutamate has been shown to be a neurotransmitter in the central nervous system of vertebrates, and it has been hypothesized that glutamate is functional as a neurotransmitter in the spinal cord dorsal horn. A monoclonal antibody to fixative-modified glutamate was used in this study to examine the light microscopic and ultrastructural profiles of glutamate-like immunoreactivity in the superficial dorsal horn of the rat spinal cord. Glutamate-like immunoreactivity was observed in neurons, fibers, and terminals of both laminae I and II. Marginal zone immunoreactive neurons ranged from 10 to 30 micron in diameter and received many nonimmunoreactive somatic synapses. In substantia gelatinosa, immunoreactive neurons were observed in both inner and outer layers, ranged 5 to 10 micron in diameter, and received few nonimmunoreactive somatic synapses. Glutamate-like immunoreactive dendrites were observed in both laminae and were contacted primarily by nonimmunoreactive synaptic terminals that generally contained small clear vesicles. Both myelinated and unmyelinated immunoreactive axons were observed in Lissauer's tract. Immunoreactive terminals contained small (40 nm) clear vesicles and generally formed simple synaptic contacts with nonimmunoreactive dendrites in laminae I and II. The results of this study corroborate the importance of glutamate as a neurotransmitter in spinal sensory mechanisms.

Ultrastructure of the central gray region (lamina X) in cat spinal cord

The central gray region (lamina X) of the lumbar spinal cord in cat was examined by electron microscopy. This region consisted of three morphological zones. Medially, the first zone was comprised of ependyma which surrounded the central canal. The ependyma in the cat spinal cord was similar to most vertebrate spinal ependyma. Secondly, a subependymal zone consisted of glial processes arranged parallel to the long axis of the spinal cord. This glial zone was widest lateral to the central canal and extended approximately 75 microns. The lateral edge of the glial zone intermingled with a neuropil zone, the third zone. The components of the neuropil zone consisted of dendrites, myelinated and unmyelinated axons, synaptic terminals, astrocytes and neurons. The dendrites and neurons generally were oriented parallel with the long axis of the spinal cord. Three synaptic terminal types were categorized according to vesicular morphology, i.e. small round vesicles, flattened vesicles and dense core vesicles. The central gray region has been implicated in nociception and has been shown to receive both primary afferent and supraspinal input. The results from this study are consistent with the central gray region being an area of multiple synaptic inputs which may form the morphological basis of nociceptive processing that ascends to brainstem nuclei.

Evidence for the inhibitory neurotransmitter gamma-aminobutyric acid in aspiny and sparsely spiny nonpyramidal neurons of the turtle dorsal cortex

In order to learn more about the anatomical substrate for gamma-aminobutyric acid (GABA)-mediated inhibition in cortical structures, the intrinsic neuronal organization of turtle dorsal cortex was studied by using Golgi impregnation, immunohistochemical localization of GABA and its synthetic enzyme glutamic acid decarboxylase (GAD), and histochemical localization of the presynaptic GABA-degrading enzyme GABA-transaminase (GABA-T). GABAergic markers are found in neurons identical in morphology and distribution to Golgi-impregnated aspiny and sparsely spiny nonpyramidal neurons with locally arborizing axons and appear to label most if not all of the nonpyramidal neurons. In addition, the GABAergic markers are found in punctate structures in a distribution characteristic of presumed inhibitory terminals. The spine-laden pyramidal neurons, the principal projecting cell type in the dorsal cortex, are devoid of labelling for GABAergic markers but are surrounded by presumed GABAergic terminals. The data complement previous physiological and ultrastructural studies that implicate aspiny and sparsely spiny nonpyramidal neurons as mediators of intrinsic inhibition of pyramidal neurons in turtle cortex. The results also suggest similarities in the functional organization of intrinsic inhibitory elements in turtle and mammalian cortex.

Organization of corticogeniculate projections in the turtle, Pseudemys scripta

The efferent pathways from the visual cortex to the dorsal lateral geniculate complex of turtles have been studied by using the orthograde and retrograde transport of horseradish peroxidase (HRP). Injections of HRP in the lateral thalamus retrogradely label neurons throughout the visual cortex. The majority of labeled neurons have somata in layer 2 of the lateral part of dorsal cortex (D2); a minority have somata in layer 3. Labeled neurons in layer 2 tend to have vertically oriented, fusiform somata and dendrites that ascend into layer 1. Labeled neurons in layer 3 have fusiform somata and dendrites, both oriented horizontally. Injections of HRP in visual cortex orthogradely label corticofugal axons. Those projecting to the lateral geniculate complex course laterally from the visual cortex, pass through the striatum (occasionally bearing varicosities), and enter the diencephalon in the ventral peduncle of the lateral forebrain bundle. Individual axons leave the ventral peduncle and run dorsally in the transverse plane, entering the dorsal lateral geniculate complex from its ventral edge. They continue dorsally, principally in the cell plate of the geniculate complex, where they bear varicosities.

Monoclonal antibodies to the turtle cortex reveal neuronal subsets, antigenic cross-reactivity with the mammalian neocortex, and forebrain structures sharing a pallial derivation

The dorsal cortex of the pond turtle (Pseudemys scripta) is a relatively simple structure consisting of two principal classes of neurons that occupy three distinct layers. Morphological, pharmacological, and physiological data suggest many similarities to the mammalian neocortex, rendering it an interesting preparation for comparative studies. We prepared monoclonal antibodies to the turtle dorsal cortex by immunizing mice with cortical tissue from adult turtles. Twelve antibodies were generated that recognize specific components of the turtle cortex. Among these, eight antibodies label only neurons and four label only ependymal glial cells. Differences in tissue staining pattern and immunoglobulin class suggest a heterogeneity of antigenic specificity among the antibodies. The staining patterns of three of our antibodies are described. TC3, like all other neuron-marking antibodies generated, labels a subset of both pyramidal and stellate cell types. It also cross-reacts with a subset of mammalian cortical neurons and labels them with a pattern similar to that observed in the turtle cortex. TC5 stains ependymal cells and their glial processes in the turtle cortex, and cross-reacts with fibrous astrocytelike processes in mammalian neocortical white matter. TC9 appears to recognize antigens of neurons sharing a pallial derivation in turtle.

The pyramidal neurons in layer III of cat primary auditory cortex (AI)

The neuronal architecture of pyramidal cells in layer III of the primary auditory cortex (AI) of adult cats was examined as a prelude to connectional and fine structural studies; in a further paper, the results of parallel studies of non-pyramidal layer III cells are presented. Layer III is about 400 micron thick, comprises about one-quarter of the thickness of AI, and lies some 400-800 micron deep to the pial surface. It is distinguished in Nissl, fiber, and Golgi preparations from layers II and IV, and also on connectional grounds, since its neurons are one of the principal inputs to the contralateral AI. Layer III may be divided into two roughly equal tiers on the basis of its neuronal and cytoarchitecture. Layer IIIa is populated by small cells with oval somata and many tiny pyramidal cells; the fiber architecture is dominated by radial bundles of medium-sized axons interspersed among columns of apical dendrites arising from deeper-lying pyramidal cells. In layer IIIb medium-sized and large pyramidal cells are more numerous, and the fiber architecture has a different, much denser texture, including extensive lateral components which invade layer IV, and large contingents of descending, probably corticofugal, axons. Five kinds of pyramidal neurons occur in Golgi preparations. Most numerous are the small, medium-sized, and large pyramidal cells; the two types of star pyramidal neurons are less common. The small pyramidal cell has a limited dendritic field and rather delicate dendrites; all but the apical one usually end in layer III. The medium-sized pyramidal cell is the most common neurons, and its rich basilar dendritic arbors are conspicuous, with their many dendritic appendages, in the layer III neuropil; their distal dendrites spread into layer IV. The largest pyramidal cells lie mainly in layer IIIb, and their lateral dendrites often mark the layer IIIb-IVa border. The apical dendrites of medium-sized and large pyramidal cells often extend to layer Ib, where they branch obliquely. The axons of these cells branch laterally after descending through layer III and toward the white matter. Often secondary or tertiary branches reascend to layer IV and more superficially; there is considerable stereotypy in this branching pattern. These numerous secondary branches contribute heavily to the layer IIIb-IVa lateral fiber plexus. The fourth variety of pyramidal cell has a round soma and a stellate dendritic field whose distal branches extend from layer V to layer I, but whose axon is chiefly in layer III. Finally, a star pyramidal cell with long lateral basilar arbors but rather smooth dendrites completes the picture.(ABSTRACT TRUNCATED AT 400 WORDS)

Quantitative light and electron microscopic analysis of cytochrome oxidase-rich zones in V II prestriate cortex of the squirrel monkey

Area 18 of V II of the prestriate cortex of the squirrel monkey was examined at both the light and electron microscopic (EM) levels for cytochrome oxidase (C.O.) activity. At the 17/18 border, the intense C.O. staining of lamina 4 abruptly ended and a new pattern continued for approximately 6 mm into the adjacent prestriate cortex. Here, periodic puffs of high C.O. activity appeared in laminae (lam.) 2 and 3, with the highest activity in lower 3 (3B) extending slightly into upper 4. There was a hint of a columnar pattern in that lam. 4 and especially 5 below the puffs were slightly more reactive than adjacent areas. A thin band of activity could also be seen in upper 5 (5A) and another one between 5 and 6. Tangential sections revealed that the puffs were arranged in alternating wide and narrow rows that radiated orthogonally from the 17/18 border. The puffs in the wider rows tended to be larger (700-1,100 micrometers in diameter) than those in the narrow rows (400-890 micrometers in diameter). The center-to- center spacing between the puffs was approximately 1,100 micrometers. Both C.O.-reactive and nonreactive stellate and pyramidal cells were found between lam. 2 and 6. Quantitatively analysis of the supragranular layers indicated that the mean area of reactive neurons was significantly larger than that of nonreactive neurons in both the puffs and interpuff (nonpuff) regions. The relative density of reactive neurons was also significantly greater than that of nonreactive neurons, and was highest within the puffs. At the EM level, reactive neurons were medium to large pyramidal cells as well as medium-sized stellates with mild to severely indented nuclei and darker cytoplasm filled with reactive mitochondria. The majority of small stellates with scanty cytoplasm and few mitochondria were nonreactive. Extensive quantitative analysis of mitochondria number and level of reactivity in different neuronal profiles indicated that the number and area of darkly reactive mitochondria was significantly higher in the puffs than in the nonpuffs, and that the majority of them resided in dentritic profiles. Between a third to half of the mitochondria in axonal profiles were darkly reactive, the frequency being slightly higher in profiles with flattened vesicles making symmetrical synapses than those with round vesicles making asymmetrical synapses. Mitochondria in axonal trunks and myelinated axons contributed to only a small percentage of the total population. Glial cells, in general, were not very reactive.(ABSTRACT TRUNCATED AT 400 WORDS)

The embryonic development of the cortical plate in reptiles: a comparative study in Emys orbicularis and Lacerta agilis

From the earliest stage of its ontogenesis, the mammalian cerebral cortex displays a remarkable cytoarchitectonic organization, with its neurons oriented radially within the cortical plate (CP). It is not known whether this radial organization of cortical neurons is characteristic of every cerebral cortex or whether it reflects a progressive phylogenetic acquisition. In order to study this question, the embryonic development of the cortex has been examined in reptiles, where it is the most primitive. Two species, Emys orbicularis and Lacerta agilis, representative of the two principal reptilian orders (chelonians and squamates), have been studied with histological methods. Golgi impregnation, and electron microscopy. Very similar patterns of cell proliferation, migration, maturation, and synaptogenesis have been observed. However, important species differences are present in the cellular organization of the cortical plate. Whereas in Emys the structure of the cortical plate is rudimentary, in Lacerta it appears well developed and quite reminiscent of its mammalian counterpart. Preliminary comparisons with embryological preparations of Sphenodon and Crocodilus niloticus show that the organization of the cortical plate displays significant variations among the different reptilian groups. The present results suggest that the radial organization of cortical neurons is not an all or nothing phenomenon but has been acquired independently and is thus a case of homoplasy, probably due to convergence (Northcutt, 81). Several possible implications of these findings are discussed and a working hypothesis based on the role of radial glial cells in the formation of cytoarchitectonic patterns (Rakic, '80) is presented.

An ultrastructural and biochemical analysis of norepinephrine-containing varicosities in the cerebral cortex of the turtle Pseudemys

The fine structure and norepinephrine content of small granular vesicle-containing profiles were studied in normal and norepinephrine-depleted cerebral cortex of the turtle, Pseudemys. The cortex was fixed for electron microscopy with the KMnO4 procedure of Koda and Bloom ('77), while the norepinephrine content was assayed wit the radioenzymatic method of Coyle and Henry ('73). Green fluorescent fibers have been described by Parent and Poitras ('74) as located almost exclusively in the outer half of the molecular layer in turtle cortex. Small granular vesicle-containing profiles are found down to 100 microns below the pial surface, but over 50% lie within 20 microns of the surface. Within the outer 100 microns of cortex, the frequency of labeled varicosities is 1.39/1,000 microns2. The average area of the norepinephrine-containing varicosities is 0.61 microns2, and there is a mean of 18.4 vesicles per single section. The average number of large plus small vesicles in an entire varicosity was estimated to be 72. Synaptic membranes are not well-preserved with KMnO4 fixation, but good examples were found of small granular vesicle-containing profiles forming both symmetrical and asymmetrical membrane differentiations. Only a small percentage of the small granular vesicle profiles were associated with a synaptic membrane differentiation in single sections. When norepinephrine-fiber synapses are seen, they usually share a postsynaptic element with another unlabeled vesicle-containing profile. Normal turtle cortex contains an average norepinephrine concentration of 1.95 micrograms/gr, which is about eight times higher than in rat cortex. The ratio of norepinephrine to dopamine is about 18 to one, suggesting that dopamine is present predominantly in a precursor pool for norepinephrine. Small granular vesicle-containing profiles were eliminated after treatment with reserpine and 6-hydroxydopamine in concentrations that were shown to reduce norepinephrine concentration by 94% and 86%, respectively. The labeled varicosities were partially depleted by midbrain hemisection and by an inhibitor of dopamine-beta-hydroxylase (FLA-63). The norepinephrine-containing varicosities are remarkably coextensive with the distribution of thalamic fibers, both in the total extent of cortex where they are found and in the depth of cortex where they terminate. The results support the idea that there is a close structural and functional association between locus coeruleus and thalamic fibers in cerebral cortex, and the apparent difference in frequency of synapses suggests that each fiber system exerts its influence on cortical cells in a different way.

The thalamocortical projection in Pseudemys turtles: a quantitative electron microscopic study

Thalamic fibers in the cortex of Pseudemys turtles were studied with the electron microscope to determine the type of synaptic vesicle they contain, the type of membrane differentiation they form, and the type of processes they contact. Following unilateral removal of the thalamus, all degenerating thalamic axon terminals are located in the outer third of the molecular layer in the rostral half of general cortex. In the middle of this zone they constitute as much as 25% of all vesicle-containing profiles. The degenerated terminals appear as electron opaque profiles, most commonly with a uniform opacity. They contain round agranular vesicles and form synapses with asymmetrical membrane differentiations. They synapse mainly on dendritic spines containing mitochondria and/or membranous sacs, although some thalamic fibers contact small clear spines, dendrites, and, rarely, cell bodies. Counts show that 86% of degenerated contacts are on dendritic spines and 14% on dendritic shafts. The spines probably all belong to the dendrites of the pyramidal cells, whose somata are located in the deep cellular layer. The dendritic shafts and somata are most likely those of the aspinous stellate neurons located in the molecular layer. Although these stellate cells are not sufficiently numerous to form a cell "layer," each transverse section through thalamic recipient cortex contains about nine of these cells and they occur in a ratio of 1:37 to pyramidal cells in the underlying main cell layer. We have calculated that in a rectangular solid of turtle cortex whose dimensions are 1 mm X 1 mm X the depth from pial surface to the underlying ventricle, there are 5.2 million thalamic fiber contacts (all in the outer 100 micrometers), 15,000 pyramidal neurons in the main cell layer, and 400 stellate cells in the molecular layer. Of the 5.2 million thalamic synapses, 0.7 million contact stellate cells and 4.5 million contact pyramidal cells. Thus each stellate cell in the molecular layer receives on the average 1,800 thalamic fiber contacts, while each pyramidal cell receives only 300 thalamic fiber synapses on the distal portion of its dendrites. The calculations lead to the conclusion that individual stellate cells receive at least six times more thalamic fiber synapses than individual pyramidal cells in turtle cortex. We suggest that the stellate cells in the thalamic input zone are inhibitory and that each thalamic volley not only excites efferent pyramidal cells but is also a powerful activator of inhibitory interneurons.

An electron microscope study of synaptic contacts in the abdominal ganglion of Aplysia californica

The fine structure of the abdominal ganglion of Aplysia californica has been studied in preparations fixed by immersion in aldehydes, either directly or after a survival of a few hours in artificial sea water. The central core of neuropil is surrounded by a rind of neuronal cell bodies floating in a subcapsular space containing a loose meshwork of neuronal and glial processes, separated by wide extracellular spaces. Large primary processes with deeply infolded membranes leave the neuronal perikarya and enter the neuropil where they branch into smaller processes containing either neurofilaments, neurotubules or both. Some have the appearance of initial segments. The neuropil is not a homogeneous structure. Rather, four types of zones can be distinguished: (1) zones of fibers of passage coursing together in the neuropil and making few synaptic contacts: (2) zones of neurosecretory fibers containing large granules and dense-core vesicles, again making few synaptic contacts: (3) zones with a great variety of synaptic contacts between medium size and small profiles; and (4) glomerular zones. The differentiated membranes of the synapses are characterized by a slight increase in density and by being regularly parallel to each other. Presynaptic densities are sometimes quite prominent but specialized dense cytoplasmic opacities have never been seen bordering the postsynaptic membranes, i.e., all synapses are of the symmetrical type. Interlemmal opacities vary considerably in density. In zone 3, the synaptic vesicles are of several sizes, are round, oval or flat, and are either clear or filled with different types of dense material. The population of vesicles within a single profile may consist either of a homogeneous group of similar vesicles or of various mixtures of two or three kinds of vesicles. In profiles with mixtures of clear and large dense-core vesicles, it is often only the clear vesicles which agglomerate towards the differentiated membranes. In such cases the large dense-core vesicles lie as a peripheral halo around the clear vesicles. Here, and especially in other large neuronal profiles not forming contact in the plane of section, they can be seen to associate specifically with mitochondria and glycogen. It is proposed that they do not contain neurotransmitters but are related to mitochondrial activities such as the storage of ATP or the movement of calcium ions. In profiles with mixtures of clear and small dense-core vesicles, both types of vesicles often touch the presynaptic membrane, suggesting the release of two transmitters or of a modulator or neurohormone with a transmitter, by a single terminal. Serial synapses are present in this zone. The glomerular zones contain small profiles forming many synaptic contacts, some of which are arranged in such a way as to suggest the existence of "reciprocal" serial synapses.