Afferents from the colliculus cortex and retina have distinct terminal morphologies in the lateral posterior thalamic nucleusкод для вставкиСкачать
THE JOURNAL OF COMPARATIVE NEUROLOGY 388:467–483 (1997) Afferents From the Colliculus, Cortex, and Retina Have Distinct Terminal Morphologies in the Lateral Posterior Thalamic Nucleus CHANGYING LING, GERALD E. SCHNEIDER, DAVID NORTHMORE, AND SONAL JHAVERI* Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 ABSTRACT We have examined the morphology of afferent endings that originate in three distinct cell groups and terminate in the lateral posterior nucleus of the thalamus (LP). Retino-LP projections were sparse, occurred throughout the nucleus, and could be classified into 1) simple en passant varicosities and terminal swellings found on poorly branched fibers in all LP subdivisions, 2) string-like configurations of varicosities detected largely in the medial subdivision of the LP, and 3) terminals resembling retinogeniculate endings occurring mainly in the rostral part of the superficial subdivision of the LP adjacent to the dorsal nucleus of the lateral geniculate body. Cortico-LP terminals fell into three classes: 1) single varicosities decorating the tips of short appendages on fine preterminal and terminal axons; 2) tiny, round varicosities studding the axon shaft; and 3) boutons of variable shape visible on mediumcaliber corticothalamic fibers. Tecto-LP terminals exhibited a large variation in morphology and density. Those found most commonly could be classified into two groups: 1) individual swellings and 2) terminal clusters arranged in a tubular configuration that enclosed a central channel, most likely occupied by the dendrite of a postsynaptic neuron. An unusual tecto-LP terminal consisted of an ovoid swelling (up to 20 µm in the long axis) from which emerged several long, thin extensions and was seen at the tips of large-diameter axons. These results show that, despite having overlapping projection zones, each set of afferents that projects to the LP elaborates terminal specializations that are structurally distinct from others projecting to the same target area. J. Comp. Neurol. 388:467–483, 1997. r 1997 Wiley-Liss, Inc. Indexing terms: axons; presynaptic terminals; rodents; visual pathway; immunocytochemistry Structure-function relationships form an essential underpinning of nervous system organization (Rouiller et al., 1986). For some regions of the brain, such as the cerebellum, hippocampus, and retina, the geometric arrangement of cell bodies, of their dendrites, and of afferent axon systems that innervate them, is known in considerable detail. However, for many central nervous system cell groups, this kind of information is fragmentary at best. Data compiled from several decades of study (using Golgi silver-impregnated material, retrogradely labeled neurons, or intracellular fills of single cells) have led to the successful definition of various cell types, but, for the most part, the delineation of afferent arbor morphology has remained refractory in all but the youngest animals. The recent introduction of anatomical tracers, such as biocytin or Phaseolus vulgaris-leucoagglutinin (PHA-L), has alleviated this lack to some degree (Gerfen and Sawchenko, 1984; r 1997 WILEY-LISS, INC. McDonald, 1992; Aggoun-Zouaoui and Innocenti, 1994; Bourassa et al., 1994; de Venecia and McMullen, 1994; Bourassa and Deschenes, 1995; Tamamaki et al., 1995), but progress in understanding the spatial configuration of axon arbors and the variations in their terminal specializations has been slow. In this report, we present observations on the morphological characteristics of afferents that project to the Grant sponsor: NIH; Grant numbers: EY05504, EY 00126, EY 02621. Changying Ling is currently at the Center for Neuroscience, University of Wisconsin, Madison, WI 53706. David Northmore is currently at the Department of Psychology, University of Delaware, Newark, DE 19716. *Correspondence to: Dr. Sonal Jhaveri, M.I.T., E25-642a, Cambridge, MA 02139. E-mail: email@example.com Received 16 August 1996; Revised 10 June 1997; Accepted 16 June 1997. 468 C. LING ET AL. lateral posterior nucleus (LP) of the hamster thalamus. This nucleus receives two major inputs, from the occipital cortex and from the superior colliculus (SC), in addition to a direct (albeit minor) innervation from the retina (Schneider, 1973; Crain and Hall, 1980a–d; BennettClarke et al., 1991). Although the rodent LP has been studied both at the light and electron microscope levels, these earlier studies have either focused on delineating the terminal zones of projections that originate in various sources or provided an ultrastructural characterization of their synaptic specializations. Little is known about the appearance of the afferent arbor or about the way afferent terminals are clustered together. In this study, we have used sensitive anterograde tracers (B fragment of cholera toxin [CT-B], PHA-L, or biotinylated dextrans [BD]) to label each of three afferent systems that innervate the LP. We report here that the terminal morphologies of axon arbors originating in each afferent cell group are distinct. Their terminal zones are largely overlapping within the LP, although the density of projection of each afferent system varies across the subdivisions of the LP. Furthermore, because we were interested in whether there is a systematic alteration in these terminal specializations following elimination of one set of afferents during early postnatal life and the subsequent sprouting of the others (Ling et al., 1997), the present findings serve as baseline anatomical observations on normal animals that have not been available to date for this nucleus. MATERIALS AND METHODS Data from 78 adult Syrian hamsters (Mesocricetus auratus) of both sexes are included in this study. Animals were purchased from Charles River Laboratories (Wilmington, MA) or bred in our colony at Massachusetts Institute of Technology (M.I.T.) and were maintained in a 14 hour/10 hour light/dark cycle. Each animal was assigned to one of three groups (see below), was deeply anesthetized with a combination of sodium pentobarbital (50 mg/kg) and valium (10 mg/kg), and was placed in a head holder. All protocols Abbreviations a CL L LGBd LGBv LP LP-c LP-d LP-m LP-s MG NOT ot PC ped PO PTA PTO R SC SGS sm SO VA-VL VB ZI alpha component of the LGBd central lateral nucleus of the thalamus lateral nucleus of the thalamus dorsal nucleus of the lateral geniculate body ventral nucleus of the lateral geniculate body lateral posterior nucleus of the thalamus caudal subdivision of LP deep subdivision of LP medialmost subdivision of LP superficial subdivision of LP medial geniculate nucleus nucleus of the optic tract optic tract paracentral nucleus of the thalamus cerebral peduncle posterior complex of the thalamus anterior pretectal nucleus olivary pretectal nucleus reticular nucleus of the thalamus superior colliculus superficial gray layer of the SC stria medullaris optic fiber layer of the SC ventral anterior-ventral lateral nucleus, thalamus ventrobasal nucleus, thalamus zona incerta used with live animals were approved by the Committee on Animal Care at M.I.T. Application of tracers Twenty hamsters in group I received an injection of 1 µl of CT-B (List Biological Laboratories, Campbell, CA; diluted to 1% in distilled water) into the vitreous chamber of the left eye via a glass micropipette attached to a microdispenser (Drummond Scientific Co., Broomall, PA). Twentytwo animals in group II received unilateral injections into the posterior cortex (for location of injection sites, see Fig. 6): Eight of these hamsters received multiple pressure injections of BD [3 µl of 5% BD (Molecular Probes, Inc. Eugene, OR) diluted in distilled water] administered via a glass micropipette (10–15 µm tip diameter) attached to a microdispenser, two hamsters received multiple injections of CT-B (3 µl total for each hamster) delivered as for the BD, and the rest of the animals (n 5 12) were each subjected to a single iontophoretic injection of PHA-L [Vector Laboratories, Burlingame, CA; 2.5% solution in phosphate-buffered saline (PBS) 0.05 M, pH 7.4] through a glass micropipette (10–15 µm tip diameter). The injection was made over a period of 15 minutes with a 5-µA positive current pulsed 7 seconds on, 7 seconds off, by a Midgard constant-current generator. Finally, 38 hamsters in group III received unilateral injections of tracers into the SC: For some of these cases, the visual cortex was suctioned out, and multiple pressure injections of CT-B (n 5 2) or of BD (n 5 12) were made into the SC under direct vision. Other group III animals received iontophoretic deposits of PHA-L (n 5 24; injection parameters as described above) after positioning the micropipette (10-µm tip diameter) in the retinorecipient zone of the SC by physiological guidance: The pipette was advanced toward the SC while monitoring the amplified, visually evoked activity, and tracer deposits were made in the region where strong multiunit activation was obtained. Four days after administration of CT-B or 8 days after BD injection, hamsters were deeply anesthetized by using sodium pentobarbital and were perfused transcardially with 0.9% sodium chloride containing 0.25% sodium nitrite as a vasodilator. This was followed by 2% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4 (PB). Animals that received PHA-L injections survived for up to 10 days and were then perfused with cold saline containing 0.25% sodium nitrite followed by 4% paraformaldehyde in 0.1 M acetate buffer, pH 6.5, then by 4% paraformaldehyde and 0.05% glutaraldehyde in 0.05 M borate buffer, pH 9.5 (details in Gerfen and Sawchenko, 1984). All brains were cryoprotected in buffered sucrose and were then cut at a thickness of 40 µm in the coronal plane on a freezing microtome. Free-floating sections were collected in buffer and were processed for localization of the tracers. Immunohistochemistry To visualize CT-B (Mikkelsen, 1992; Angelucci et al., 1996), tissue was thoroughly washed in PBS and then immersed sequentially in 0.1 M glycine (made up in PBS) for 30 minutes; in 2% Triton X-100, 2% normal rabbit serum (NRS) plus 2.5% bovine serum albumin (BSA) for 1 hour at room temperature; in goat anti-CT-B antibody (List Biological Laboratories; diluted 1:8,000 in PBS containing 2% Triton X-100, 2% NRS, and 2.5% BSA) for 96 hours at 4°C; in biotinylated rabbit anti-goat secondary antibody (Vector Laboratories; diluted 1:200 in PBS con- AFFERENTS TO LATERAL POSTERIOR NUCLEUS taining 2% Triton X-100, 2% NRS, and 2.5% BSA) for 90 minutes at room temperature; and in an avidin-biotinperoxidase complex (Elite ABC kit; Vector Laboratories) for 90 minutes at room temperature. The horseradish peroxidase (HRP) was visualized histochemically by using 3,38-diaminobenzidine tetrahydrochloride (DAB) as a chromagen, and the reaction product was enhanced with the addition of 0.02% cobalt acetate. To visualize PHA-L (Gerfen and Sawchenko, 1984), sections were incubated overnight at 4°C in PBS containing 0.3% Triton X-100 and 2% NRS; then in goat anti-PHA-L1E antibody (Vector Laboratories; diluted 1:1,000 in PBS containing 0.3% Triton X-100 and 2% NRS) for 96 hours at 4°C; then in rabbit anti-goat biotinylated secondary antibody, followed by avidin-biotin-HRP, and, finally, the HRP was visualized with DAB, as described above for the CT-B-labeled sections. To visualize BD (Rajakumar et al., 1993), tissue was reacted directly with avidin-biotin-peroxidase complex (Elite ABC kit; Vector Laboratories; diluted 1:100 in PB) for 90 minutes at room temperature. After washing thoroughly, sections were reacted with DAB, as described above. All sections were mounted on chrome alum- and gelatincoated slides, air dried, dehydrated, and coverslipped. Tissue was analyzed under the microscope at both low (3100) and high (31,000) magnification and was photographed, or axons were drawn with the aid of a drawing tube attachment. Subdividing LP The cytoarchitecture and fiber architecture of the hamster LP was examined in sections from our laboratory collection of normal adult brain sections processed with silver stains (Schneider, 1969) for viewing normal fibers, with the Loyez stain for the distribution of myelinated axons, and with cresyl violet for studying the distribution of cell bodies. In addition, borders of individual subdivisions were also determined on the basis of hodological observations for this region (Schneider, 1969, 1973, 1975; Lent, 1982; Jhaveri et al., 1985; Ling et al., unpublished observations). RESULTS Architecture of the hamster LP Our subdivisions of the hamster LP are slightly different (see Fig. 1) from those described by Crain and Hall (1980a). We defined a superficial subdivision, referred to as LP-s, that was characterized by lighter neuropil than in the rest of the nucleus, as seen in fiber stains. This region received very dense projections from the superficial layers of the SC. At caudal levels of the nucleus, LP-s extended from the edge of the LGBd (laterally) to the pretectum (medially), whereas, in more rostral sections, LP-s was abutted medially by the dorsal extension of the deep subdivision (LP-d), which reaches the diencephalic surface at this level. In normal fiber stains, LP-d had a denser neuropil, and, through it, traveled many corticofugal fibers of passage destined for pretectal and tectal targets. In silver-degeneration material, LP-d was seen to receive input from deeper tectal layers as well as from visual cortex. Although LP-s and LP-d are fairly similar in extent to what Crain and Hall call LP-rl and LP-rm, we prefer the ‘‘superficial’’ and ‘‘deep’’ nomenclature, because it allows orientation of the subdivisions relative to the optic tract; 469 see Discussion. At rostral levels in the hamster, the a subdivision of the LGBd was seen to be continuous with LP-d. At most levels of the LP, an additional, medial-most region of lighter neuropil was evident; we designated it as LP-m. In sections immunostained with an antibody against GAP-43, this subdivision stands out by virtue of its denser immunoreactivity (Moya et al., 1990; Schneider and Benowitz, unpublished observation). In some sections, this lighter neuropil extended more superficially, but such a delineation was not seen consistently, and is not illustrated in Figure 1. Caudally, LP-d extended into a region of dense fibers: This subdivision was referred to as LP-c by Crain and Hall (1980a) and likewise by us (however, see Discussion). In Nissl stains, these subdivisions were difficult to differentiate in the hamster. Ventral to LP-d, the boundary with the underlying posterior or ventral thalamic nuclei was not a sharp one. It was drawn by convention along the course of the external medullary lamina, although some projections from the tectum did extend below this line. Morphological specializations of retino-LP projections Injections of CT-B into the vitreous chamber of the eye result in uptake and anterograde transport of the tracer by retinal ganglion cells. CT-B-labeled axons were observed in all known retinorecipient zones, and sparse, labeled fibers were also revealed in some regions of the brain that have not been recognized previously as targets for primary visual afferents (Ling et al., 1994). Earlier reports have documented that a restricted portion of LP-s adjacent to the superficial optic tract receives a small focus of contralaterally projecting retinal fibers (Schneider, 1973; Crain and Hall, 1980a). Labeling with CT-B revealed that, in addition to these tiny foci, scattered retinal terminals were also present throughout much of the rest of the LP; however, the density of the retino-LP projection was higher in the LP-s at rostral levels (Fig. 2a) and in the LP-m at central levels of the nucleus (Fig. 2b). Retinal axons that entered the LP left the main optic tract just below the ventral nucleus of the lateral geniculate body (LGBv) and followed a deeper trajectory through the parenchyma of the LGBv and LGBd, contributing to fibers of the internal optic tract (IOT; Schneider and Jhaveri, 1983). These axons penetrated the LP across its border with the LGBd and, for the most part, were bundled into distinct fascicles that coursed parallel to the surface of the dorsolateral thalamus (Fig. 2a). (It is possible that some retinal axon collaterals also entered the nucleus directly from the superficial optic tract; such a collateralization, if it did occur, would be difficult to detect in our material because of the dense labeling of surface optic tract axons.) Most retinal fibers that passed through the LP exhibited sparse ramification, and their terminal specializations could be classified into three groups. Varicosities and simple endings on individual collaterals. A scattered distribution of fine fiber collaterals was noted emerging from retinal axons as they coursed through the LP. These afferents bore en passant swellings of various sizes: tiny (,1 µm in diameter; Fig. 3a), small (1–2 µm in diameter; Fig. 3b), and medium-sized (.2 µm in diameter; Fig. 3c) varicosities were loosely strung along the length of the fiber collaterals. A simple terminal bouton was often noted at the end of each collateral of the preterminal axon (Fig. 3, arrowheads). No specific clustering of these swellings was apparent. Such ramifications 470 C. LING ET AL. Fig. 1. Top: Schematic diagram illustrating the subdivisions of the lateral posterior nucleus (LP) in a series of frontal sections taken at rostral (left), middle, and caudal (right) levels of the nucleus. The superficial subdivision (LP-s), deep subdivision (LP-d), medial subdivision (LP-m), and caudal subdivision (LP-c) are indicated. At caudal levels of the nucleus, LP-s extends from the edge of the dorsal nucleus of the lateral geniculate body (LGBd; laterally) to the pretectum (medially), whereas, in more rostral sections, it is abutted medially by the dorsal extension of LP-d. LP-d is continuous with the alpha (a) subdivision of LGBd at rostral levels and is replaced by LP-c at caudal levels of the nucleus. LP-m forms the medialmost subdivision and is visible at most levels. Bottom: Photomicrographs from silver-stained frontal sections through the LP showing the fiber architecture through the nucleus in relation to its subdivisions. The orientation and level of each section is indicated by the gray box in the corresponding schematic diagram above each section. For abbreviations, see list. Scale bar 5 1,000 µm for top, 200 µm for bottom. were found in all subdivisions of the LP, with a slightly higher density in LP-m at central levels of the nucleus (Fig. 2b). String-like complexes. These were found primarily in LP-m at midlevels of the nucleus (Fig. 2b), and their distribution overlapped that of the simple endings described above. Each complex was comprised of several (at least two) individual fibers that were studded with boutons of various sizes, primarily of the small and medium category (Fig. 4). The swellings occurred closer together than the simple endings described above. A longitudinal arrangement of such bouton-studded fiber strings could frequently be visualized in which two or more fibers traveled together for a distance and then separated (Fig. 4). Terminals resembling normal retinogeniculate endings. Previous studies (Erzurumlu et al., 1988) have identified at least two distinct types of retinal terminals in the LGBd (labeled terminals shown in Fig. 5a–d are from the LGBd). Type R1 terminals consist of large, oval swellings that are strung along thick axons (Fig. 5a), whereas type R2 terminals are comprised of rosette-like clusters of small or medium varicosities that emerge from fine-caliber retinal afferents (Fig. 5b–d). In general, more than one afferent collateral contributes to each cluster of R2s (Fig. 5c,d). (Ubiquitous type R3 retinogeniculate terminals are comprised of endings that do not fall into the type R1 or type R2 categories, and they are not discussed here.) Both of these classes of endings could be identified in the LP: Terminals with morphologies typical of type R1 (Fig. 5e,f) and R2 retinogeniculate endings (Fig. 5g–j) were found primarily in LP-s at rostral levels. They were present near the border of the LP and LGBd but were distinctly outside the latter cell group, in a region designated as the retinorecipient zone of the LP in earlier studies (Schneider, 1973; Crain and Hall, 1980a). Terminals of visual cortico-LP afferents The rodent visual cortex is comprised of three distinct architectonic fields (Fig. 6): striate cortex or area 17 and a peristriate ‘‘belt’’ formed by area 18a laterally and 18b medially (Schneider, 1969; Caviness, 1975; Tiao and Blakemore, 1976a,b; Lent, 1982; Olavarria and Van Sluyters, 1985; Rhoades et al., 1987; Lewis and Olavarria, 1995). In AFFERENTS TO LATERAL POSTERIOR NUCLEUS 471 Fig. 2. Photomicrographs of frontal sections through the LP of adult hamsters that had injections of cholera toxin in the contralateral eye (a,b), of Phaseolus vulgaris-leucoagglutinin (PHA-L) in the ipsilateral visual cortex (c), and of PHA-L in the ipsilateral superior colliculus (d). These views indicate the relative densities and distributions of inputs from these three cell groups. In a and b, sparse retino-LP terminals extend from optic axons that run through the nucleus parallel to the surface of the thalamus. Retino-LP terminals have a relatively high density in LP-s at rostral levels (a) and in LP-m at mid-LP levels (b). In c, cortico-LP projections at mid-LP levels have a high density in LP-d, whereas, in d, tecto-LP projections at mid-LP levels show a dense projection in LP-s. Arrowheads point to the border between the LP and LGBd in each micrograph. For abbreviations, see list. Scale bar 5 50 µm in a, 100 µm in b–d. our study, cortical injections were largely restricted to area 17 (for the location of the centers of injection sites, see Fig. 6). Each deposit of tracer (PHA-L, CT-B, or BD) resulted in a cylindrical-shaped region of labeled cells and processes approximately 0.6–1.0 mm in diameter and 0.6–1.0 mm in depth. The injection site extended through most or all cortical layers (see Fig. 7 for typical injection site). CT-B and BD were transported both anterogradely and retrogradely, whereas only anterogradely filled axonal profiles that projected to the LP were seen following iontophoretic injections of PHA-L. Results that were confirmed with all three methods are described here, with photographic documentation derived from the PHA-L-labeled tissue. Labeled corticofugal fibers en route to the LP coursed largely in the external medullary lamina, peeling off laterally to innervate target cells. Terminal arbors of cortico-LP axons were observed throughout the LP on the side of the injection, including the LP-s, a region that also receives a dense projection from the SC (Fig. 2c; see below). On morphological grounds, cortico-LP terminals could be classified into three types. The first two types occurred on fine fibers (diameter 1 µm or less), which constituted the majority of cortical arbors. They bore tiny, round varicosities (diameter ,1–1.5 µm; Figs. 8, 9), and the density of short branches and spine-like appendages varied signifi- cantly from one fiber to the next (Figs. 8, 9a–c). According to the morphological characterization by Bourassa and Deschenes (1995) for cortico-LP axons in the rat, these fibers could be divided into two groups: 1) Type I afferents were branched and bore numerous short appendages (Fig. 8, arrow, Fig. 9c), and 2) type II fibers were less heavily branched but were studded with varicosities along the length of the shaft or at the tip of the ramification (Fig. 8, arrowhead, Fig. 9a). Although some labeled fibers could be classified as belonging to one or the other of these types, intermediate forms, like that shown in Figure 9b, were also seen. Thus, in the hamster, these two groups of fibers may represent extreme variations within a single category (see Discussion). Long lengths (up to 200 µm) of preterminal arbors could be traced in 40-µm-thick coronal sections, suggesting that these arbor ramifications occurred primarily in the transverse plane (Fig. 8). 3) In addition, a less frequently seen fiber type (type III) consisted of mediumsized axons (,2 µm diameter). Type III fibers bore mediumsized varicosities (3–4 µm diameter; Fig. 9d,e) that had a sparser distribution along the length of axons. Although points of bifurcation could be detected (Fig. 9e, arrows), by and large these axons appeared to be unbranched within the LP. The projection zones of the three types of cortico-LP 472 C. LING ET AL. Fig. 3. Photomicrographs showing higher magnification views of cholera toxin-labeled varicosities and simple endings on individual retino-LP axon collaterals. a: Tiny boutons (,1 µm average diameter). b: Small boutons (1–2 µm average diameter). c: Medium boutons (.2 µm average diameter). Arrowheads point to terminal boutons. Scale bar 5 10 µm. Fig. 4. Photomicrographs showing the detailed morphology of string-like complexes formed by cholera toxin-labeled retinal axons in LP. Each complex is comprised of segments of more than two collaterals that are densely studded with boutons of small (a), medium (b,c), and tiny (d) size. Scale bar 5 10 µm. arbors were largely overlapping (Figs. 8, 9d). At a qualitative level, fiber types I and II, bearing tiny varicosities, predominated in most regions of the LP, with fewer medium-sized axons bearing simple varicosities. This was true even in the anterolateral corner of the LP-d, where the heaviest density of medium-sized cortical afferents was seen. Terminals of tecto-LP afferents In most of our cases, tracer injections were restricted to the upper half of the SC, including the superficial gray (SGS), the optic fiber layer (SO), and the intermediate gray layer (a typical injection site is shown in Fig. 7b), although, in a few (less than 1⁄3) cases, injection sites also encroached AFFERENTS TO LATERAL POSTERIOR NUCLEUS Fig. 5. Detailed morphology of type R1 and type R2 retinogeniculate terminals (a–d), and type R1- and type R2-like retino-LP terminals (e–j) seen in animals in which retinal axons are labeled with the B fragment of cholera toxin (CT-B). In the dorsal nucleus of the lateral geniculate body (LGBd), type R1 terminals form large, ovoid boutons Fig. 6. Schematic diagram representing a dorsal view of the caudal portion of the hamster brain indicating the center of placement of PHA-L (circles), biotinylated dextran (black dots), and cholera toxin (gray dots) injections on the right side in relation to subdivisions of the occipital cortex, which are delineated on the left side (17, cortical area 17 or striate cortex; 18a, cortical area 18a or lateral juxtastriate cortex; 18; cortical area 18, medial juxtastriate cortex or cortical area 18b). Scale bar 5 1,500 µm. upon deeper layers, including the dorsal edge of the central gray. With all three tracers, the same colliculo-LP terminal types were identified, and no retrogradely labeled cells 473 that are loosely strung along thick fibers (a), whereas type R2 terminals consist of medium- and small-sized boutons that form rosette-like clusters and that emerge from fine-caliber axons (b–d). R1-like retino-LP terminals (e,f) and R2-like retino-LP terminals (g–j) resemble the endings in the LGBd. Scale bar 5 10 µm. were seen in the LP (the LP does not normally project to the tectum). No attempt was made in this study to separate terminal types according to whether they originated in the SGS, SO, or deeper layers (earlier studies have shown that tecto-LP projections originate largely from the lower portion of the SGS and the SO; albino rat: Sugita et al., 1983; other mammals: for review, see Huerta and Harting, 1984; hamster: Mooney et al., 1988; BennettClarke et al., 1991). Our results showed that the SC projects bilaterally to the LP, with the ipsilateral pathway being by far the predominant one. Tectal afferents to the ipsilateral LP entered the nucleus through the parenchyma of the lateral pretectum or via the brachium of the SC, whereas axons that projected contralaterally crossed in the postoptic commissure and ascended within the opposite optic tract. On the ipsilateral side, the highest density of tectal projections occurred in the LP-s (Fig. 2d), but labeled afferents were also found throughout the rest of this thalamic target. There was a large variation in the morphology of tecto-LP terminals, but the most commonly seen endings could be grouped into two types: 1) Individual terminals were found on fibers that had an average diameter of about 1 µm. These afferents were smooth for the most part and bore a sparse distribution of swellings that varied in diameter from 1.5 to 4 µm. The majority of such swellings were detected on short appendages that decorated axon trunks; some were located at the tip of a 474 C. LING ET AL. Fig. 8. Photomicrographs of PHA-L-labeled cortico-LP afferents. The majority of cortico-LP projections are formed by fine-caliber axons bearing tiny, round varicosities. Many of the boutons on these axons decorate the tips of short appendages along the length of the fiber (type I axon, arrow; see also Fig. 9), but varicosities are also present along the length of the axon shaft (type II axons, arrowhead). Considerable variation in the density of varicosities is seen from one collateral to the next. Scale bar 5 25 µm. Fig. 7. Photomicrographs illustrating typical injection sites in area 17 (a) and in the superior colliculus (SC; b). Arrow in a indicates the ventral border of layer VI. In both cases illustrated here, the tracer used was PHA-L. Dashed line in b indicates the border between the superficial gray layer of the SC and the optic fiber layer; arrow indicates the tectal midline (b). Scale bar 5 500 µm. terminal branch (Fig. 10). Occasionally, several collaterals bearing these types of endings were seen to form loose clusters (Fig. 10, long arrows). These individual terminals were morphologically distinct from the simple endings on retino-LP afferents by virtue of their more frequent branching and by the presence of short appendages along the axon shaft (Fig. 10, short arrows). 2) Tubular clusters of tecto-LP terminals were each comprised of 20–40 boutons aligned around a central, cylindrical channel that was approximately 1–3 µm in diameter and up to 20–30 µm in length (Fig. 11b,c, arrows). Boutons within each cluster were so closely aggregated at times that it was difficult to delineate individual swellings (Fig. 11a,d, arrows). The central channel had a size and shape consistent with the possibility that it was occupied by the dendrite of a postsynaptic neuron (see Discussion). The average outer diameter for a single tubular cluster was about 4–6 µm. Other groups of boutons, which may or may not have surrounded a central cavity of a similar dimension, formed a round or ovoid (rather than longitudinal) arrangement (Fig. 11, arrowheads). We assume these arose as a result of transverse or oblique sections through the tubular clusters. Whereas the individual endings had a dense distribution, the tubular clusters were more visually striking because of their distinctive morphology (Fig. 2d). In addition to the two types of tecto-LP endings described above, a third, infrequently seen terminal occurred on largediameter (2–4 µm) axons and had a unique morphology. It consisted of a large, ovoid terminal swelling (up to 20 µm) at the axon tip from which emerged several long, thin extensions (Fig. 12). Terminals with this kind of morphol- ogy could also be smaller in size and occurred on thinner axons. The density of the different tectal efferents varied in different regions of the LP, but both of the commonly seen types of terminals of axons from SC were present throughout this target, including the LP-d, the region that was previously believed to receive a projection only from visual cortex (LP-rm in Crain and Hall, 1980a). The distribution of the various classes of tecto-LP endings was largely overlapping. The highest density of tecto-LP projections occurred in the LP-s, a region that also received a moderate projection from the contralateral SC and from the ipsilateral visual cortex (area 17), in addition to a sparse projection from the contralateral retina (Fig. 13). Labeled terminal profiles shown in Figures 3–5 and 8–11 were derived primarily from this region. (In Fig. 4, the stringlike complexes of retinal origin shown are in the LP-m.) DISCUSSION The LP (or portions of the pulvinar of cats and monkeys) is known to participate in an extrageniculate visual pathway that links the SC and the pretectum to visual cortex in mammals (Diamond, 1973; Graybiel and Berson, 1980, 1981). It is also widely accepted that the LP receives inputs at least from the SC, the visual cortex, and the retina and that it sends afferents to the visual cortex (Nauta and Bucher, 1954; Abplanalp, 1970; Benevento and Ebner, 1970; Harting et al., 1973; Karamanlidis and Giolli, 1977; Robson and Hall, 1977a,b; Arango and Scalia, 1978; Gould et al., 1978; Dursteler et al., 1979; Hollander et al., 1979; Karamanlidis et al., 1979; Caviness and Frost, 1980; Coleman and Clerici, 1980, 1981; Crain and Hall, 1980a; Perry, 1980; Linden and Rocha-Miranda, 1981; Raczkowski and Diamond, 1981; Lent, 1982). A considerable amount of information is available about the distribution Fig. 9. Detailed morphology of cortico-LP terminals. The column on the left (b–e) shows camera lucida drawings of axons from the corresponding photomicrographs in the right column. a: Type II axon with varicosities primarily along the axon shaft. b: Axon arbor with morphology that is of a form intermediate between type I and type II axons. c: Type I axon arbor—note the numerous short appendages studded with tiny varicosities. d: Illustration of the overlapping distribution of the various terminal types. e: Type III axon arbors; arrows indicate points of bifurcation on the type III arbors. Scale bar 5 10 µm. 476 C. LING ET AL. major input from the external nucleus of the inferior colliculus (unpublished data) and could just as well be considered to be part of the medial geniculate body, as has been noted for rats (LeDoux et al., 1985; GonzalezHernandez et al., 1991). In terms of projection zones, our observations confirm and extend those of Crain and Hall. We show that afferents to the LP have a significantly more widespread projection zone than was detected previously and that the overall zones of termination from the retina, SC, and visual cortex are overlapping, including in the LP-d (Fig. 2), a region that was thought previously to receive only visual cortical afferents. However, the relative density of each afferent system varies within the subdivisions of the LP (Fig. 13). Differences between our results and those of Crain and Hall are most likely due to the greater sensitivity of the anterograde tracers we have employed for this study and the greater facility for detailed sampling over the entire nucleus with the light microscope. Furthermore, staining for degenerating axons or using 3H-amino acid autoradiography results in a discontinuous labeling of axons, with varying levels of background staining. Thus, with these older tracing techniques, it would have been difficult to differentiate a sparse terminal projection (such as that from the retina to the LP) from labeled axons of passage. In the present study, we have used several different tracers, each of which provides a diffuse fill of preterminal axons and their terminal arbors, making it easier to distinguish sparse terminal specializations from axons that are passing through the target region. Differential morphological specification of afferent arbors from different sources Fig. 10. Camera lucida drawings of individual tecto-LP terminals. These endings are of various sizes, are sparsely distributed, and are seen on fine-caliber axons. Long arrows, loose bouton clusters; short arrows, short appendages along axon shafts. Scale bar 5 10 µm. and ultrastructural characteristics of LP terminals that derive from different sources. Despite the large volume of anatomical work devoted to the study of this nucleus, the present results are novel in providing light microscopic information on the morphology of terminal specializations on afferents arising from the retina, cortex, or tectum and projecting to the LP. Overlapping terminal fields of axons from different sources Crain and Hall (1980a) delineate three subdivisions of the hamster LP, basing their conclusions on cytoarchitectonic and myeloarchitectonic criteria and on a description of the projection pattern of retino-LP, cortical-LP, and tecto-LP afferents to each of these subdivisions. Our subdivisions agree in part with this earlier definition, but we have modified the borders slightly on the basis of additional data from normal fiber stains and from hodological studies. First, to simplify the nomenclature, we have referred to the two major subdivisions as superficial and deep in relation to the optic tract. Thus, our LP-s is essentially the same as LP-rl of Crain and Hall (1980a). In addition, primarily on the basis of normal fiber architecture, we have distinguished a medial subdivision (LP-m) that was not delineated in the earlier study. Our definition of the borders of the LP-c agree with those of Crain and Hall, but we point out that the hampster LP-c receives a Our results document the morphology of afferent arbors in the LP, with a special focus on the form and clustering of terminal endings, and they provide a perspective that is not generally possible to obtain from ultrastructural observations. We show that afferent arbors and terminals in this nucleus display source-specific features that differentiate them from other afferents projecting to the same zones. Afferents from the retina. Retinal axons headed toward the SC follow one of two routes: Most of them course along the diencephalic surface in the superficial optic tract. However, at the ventral edge of the LGBv, significant numbers of retinal axons diverge from the surface fibers and enter the internal optic tract (Schneider and Jhaveri, 1983; Bhide and Frost, 1991). These fibers course through the parenchyma of the LGBv, LGBd, and LP on their way to caudal targets. In earlier studies that used degeneration techniques (or anterograde transport of 3H-amino acid, or HRP) to label retinal axons, it was concluded that, with the exception of a few tiny foci of terminals, most of the labeled retinal axons in the LP were merely passing through, and there was virtually no direct retino-LP projection (Schneider, 1973; Crain and Hall, 1980a). An exception to this was thought to occur during development (Perry and Cowey, 1982). It was argued on the basis of HRP labeling that there could exist a transient retino-LP projection that gets eliminated during early postnatal life. Those authors suggested that the abnormal retinal projection to the LP that forms after early ablations of the tectum (Schneider, 1973) results from an arrest of this elimination process. Although irrevocable evidence for such a transient retino-LP projection during development has not been forthcoming, the use of CT-B to label retinal axons in the AFFERENTS TO LATERAL POSTERIOR NUCLEUS Fig. 11. a–f: Photomicrographs (a,c–f) and camera lucida drawings (b) showing the morphology of tecto-LP terminals that form tubular clusters. A typical tubular cluster (arrows) is comprised of 20–40 boutons, which form an array that can extend up to 30–40 µm in length and which are aligned around a central channel (1–3 µm in 477 diameter). Other boutonal clusters (arrowheads) form round or ovoid configurations and also surround a central cavity of a similar dimension—these are most likely cross-sections through the tubular clusters. Because of the dense packing of the terminals, some individual varicosities are not clearly visualized (a,d). Scale bar 5 10 µm. 478 Fig. 12. Camera lucida drawings (top) and photomicrograph (bottom) of PHA-L-labeled superior colliculus (SC)-LP terminals found on large-caliber fibers. The endings consist of a large ovoid swelling at the axon tip with several long, thin extensions emerging from it. Such endings are infrequently seen in the tecto-LP projection. Scale bar 5 10 µm. developing animal may yet provide the definitive answer. Clearly, in the adult, varicosities are visible on retinal fibers as they course through the LP. However, it is not known whether some of these specializations form on axons that are collaterals of fibers that go on to other targets or whether they are retinal axons that are destined specifically for the LP. The close resemblance between type R1 and type R2 retinal terminals in the LGBd and those in the LP might suggest that some retino-LP axons are a continuation of retino-LGBd afferents. In any case, our observations clearly document that terminal and preterminal specializations on retinal axons are scattered throughout the LP and are not restricted to the tiny, dense foci described in the prior studies. Although, overall, this projection is relatively minor, the widespread nature of C. LING ET AL. optic afferents in most LP subdivisions is relevant to the distribution of sprouted retino-LP terminals that occur following early tectal lesions (see Ling et al., 1997). Ultrastructural correlates of the swellings and terminal specializations we illustrate here remain unclear. Only one type of synaptic bouton, the RS terminal, has been described in the LP as deriving from the retina. However, on the basis of size and clustering, it is not possible to assign the RS terminal to any one of the retino-LP specializations that we have observed in this study. Our type R1 endings are too large, and one would expect most of the type R2 endings to be clustered in the EM, as they are in the LGBd (Szentagothai, 1963; Szentagothai et al., 1966; Guillery, 1969; Guillery and Scott, 1971; Famiglietti and Peters, 1972; Lund and Cunningham 1972; Rafols and Valverde, 1973); with regard to the swellings that are on the string-like complexes or that comprise the simple endings on individual collaterals, these varicosities exhibit a substantial variation in size, and they cannot be allotted unequivocally to correspondence with the RS terminals at the ultrastructural level. Thus, our data indicate that a more detailed ultrastructural study of the LP, perhaps in combination with CT-B labeling of retinal afferents, is necessary to understand how and where the retinal terminals contact postsynaptic targets in the LP. Afferents from the SC. The photographic documentation of the tecto-LP axon terminals shown in the present study was derived from material labeled with PHA-L, but it was confirmed in cases wherein BD was also used as the anterograde tracer. There is considerable evidence that PHA-L is transported long distances along axons in the anterograde direction (see, e.g., Gerfen and Sawchenko, 1984, 1985; Carlson and Heimer, 1986; Grove et al., 1986; Rockland et al., 1995). However, the possibility that at least some of the labeled profiles we see result from retrograde transport of retinotectal axons and the possibility that our observations are confounded by the labeling of collateral branches of these back-filled cells into LP must be considered. Complete labeling of neuronal cell bodies and dendrites was detected after PHA-L injections, but only in the region of the injection site. In addition, longdistance retrograde labeling by PHA-L is occasionally reported, but the labeling density is usually conspicuously weaker, and the reaction product is granular, not diffuse as shown in our material (Kita and Kitai, 1987; Lee et al., 1988; Shu and Peterson, 1988; Clarke et al., 1993). Furthermore, such labeling is reportedly obtained with the use of larger pipette tips for making the iontophoretic injections than those used in this study or when making tracer deposits with pressure injections rather than electrophoretically, as we have done (Gerfen and Sawchenko, 1984, 1985). Based on these considerations, we are confident that the labeled terminals shown in this study result from anterograde transport of the tracer. A tecto-LP projection has been identified in a wide variety of mammals (Altman and Carpenter, 1961; Morest, 1965; Tarlov and Moore, 1966; Martin, 1969; Abplanalp, 1970; Hall and Ebner, 1970; Niimi et al., 1970; Rafols and Matzke, 1970; Mathers, 1971; Casagrande et al., 1972; Graybiel, 1972; Harting et al., 1973; Benevento and Fallon, 1975; Glendenning et al., 1975; Mooney et al., 1988). Although there are notable variations in the relative size and subdivisions of the LP among different species, there are also many similarities. In general, tecto-LP projections arise largely from the superficial, retinorecipient layers of AFFERENTS TO LATERAL POSTERIOR NUCLEUS 479 Fig. 13. Schematic diagram summarizing the results on the density and distribution of the three systems of afferent arbors that originate from different sources and terminate in the LP. Retino-LP terminals: dots, type R1-like and type R2-like terminals; bars, stringlike configurations of retino-LP terminals; squares, varicosities and simple endings on retino-LP collaterals. Cortico-LP terminals: small squares, type I and II arbors; large squares: type III arbors. SC-LP terminals: bars, tubular clusters; circles, individual terminals. For abbreviation, see list. the SC, with ipsilateral afferents predominating. Ultrastructural study of tecto-LP terminals reveals three types of synaptic endings, referred to as M terminals, RS termi- nals, and F terminals (Robson and Hall, 1977a; Crain and Hall, 1980a). In hamsters, the majority of tecto-LP axon arbors form M terminals, which are of medium size, 480 contain round vesicles, and are generally found tightly packed together around the shaft of a large proximal dendrite of a postsynaptic neuron. To a lesser extent, tectal afferents also form RS and F terminals in the LP: RS terminals are small sized (,1 µm in diameter), contain round vesicles, generally synapse on the shaft of a small dendrite, and are located outside the large synaptic clusters dominated by the M terminals. F terminals, so called because of their flattened or pleomorphic synaptic vesicles, are intermediate in size between the M and RS terminals and infrequently participate in the large synaptic clusters. Our light microscopic observations on tecto-LP projections can be partially correlated with these ultrastructural data. Most likely, the M terminals described by Crain and Hall are the ultrastructural equivalent of what we refer to as the tubular clusters that surround a central channel— swellings that comprise these clusters are of a size that corresponds well with that of the medium-sized M terminals. In the electron microscope (EM), boutons of M terminals are seen closely abutted to each other, which could explain why, in some of our examples of tecto-LP clusters, individual varicosities could not be delineated at the level of the light microscope (see Fig. 11). The average outer diameter for a tubular cluster in our material was about 4–6 µm, the same as that measured for clusters of M terminals in the figures of Crain and Hall (see, e.g., Fig. 6 in Crain and Hall, 1980a). The tubular conformation of these clusters is consistent with the hypothesis that the endings are closely aligned along the proximal dendrites of LP neurons. Variations in size of the simple endings that we describe here make it possible that at least some of them correspond to the RS terminals of Crain and Hall. However, if size was the only criterion, then the simple endings most closely match the category of terminals referred to as the F terminals. Nonetheless, it should be underscored that the tecto-LP projection is complex, and we have described here the morphology of only the most commonly found endings. Regarding the large endings with trailing processes (Fig. 12), no EM equivalent of these has been reported in the literature; however, they are rare enough, and one explanation for such an omission might be the difficulty of obtaining a thorough ultrastructural sampling in all parts of the nucleus. These terminals bear similarities to large endings in the medial nucleus of the trapezoid body (cf. Morest, 1968, 1975). Projections from the visual cortex. It has been documented for many mammalian species that the LP receives a major projection from the ipsilateral visual cortex, including area 17, and has a reciprocal projection back to the visual cortex (Updyke, 1975; Robson and Hall, 1977a,b; Giolli et al., 1978; Ogren and Hendrickson, 1979; Crain and Hall, 1980a; Lent, 1982; Jones, 1985; Klein et al., 1986; Bourassa and Deschenes, 1995). In the rat, terminals of cortico-LP axons are of at least two types and likely originate in neurons of cortical lamina V as well as lamina VI (Bourassa and Deschenes, 1995). In hamsters, morphologically distinct cortico-LP synapses have been described previously at the ultrastructural level: 1) RL terminals are large (.2-µm-long axis), contain round vesicles, and synapse on appendages of proximal dendrites of LP neurons; they do not show a tendency to form clusters. RS and F terminals have the same characteristics as those described for the tecto-LP projection (Crain and Hall, 1980a). Our light microscope C. LING ET AL. characterization of cortico-LP terminals showed that the majority of them were tiny, round varicosities (1–1.5 µm in diameter; Fig. 9a–d) located along the shaft of fine axons, but they were also found on the tips of short appendages that emerged more or less perpendicularly from the parent axon. Less frequently, medium-sized boutons (3–4 µm diameter; Fig. 9d,e) were found that had a sparser distribution along the length of medium axons. The larger size of the RL terminals precludes them from corresponding to the tiny swellings we have observed on cortico-LP arbors; this suggests that the medium-sized cortico-LP endings would be candidates for the light microscope correlate of the RL terminals. 2) On the other hand, the most common of cortico-LP terminals, those on the spiny fibers, match the description of the RS terminals of Crain and Hall (1980a) on the basis of size alone. However, there is a discrepancy between their study and ours, in that we find that the tiny endings on fine fibers are the most commonly labeled cortico-LP terminals, whereas, in the EM, the majority of cortico-LP terminals seen are of the RL variety (and size considerations disallow any correspondence between the tiny varicosities and the RL terminals). One possibility is that the difference in the two studies arises from variations in the involvement of lamina V vs. lamina VI of the cortex at the injection site (see Bourassa and Deschenes, 1995). In addition, we have attempted to restrict our injection sites to area 17, whereas, in the earlier study, tracer applications spread also into area 18 and posterior temporal cortex (Crain and Hall, 1980a). Thus, many of the RL terminals described in the EM work might originate in the cortex outside of area 17. However, it should be underscored that attempts at making such correlations between light microscopic and ultrastructural observations, including the ones discussed above, are confounded; thus, in the EM work, only swellings that have synaptic vesicles associated with membrane specializations are ascribed to specific terminal classifications, whereas no differentiation between swellings that do or do not make synaptic connections is possible in the light microscope observations. Bourassa and Deschenes (1995) injected cells in cortical lamina V or lamina VI, traced the projections to LP, and found that the terminal ramifications of axons from each lamina were distinct: All corticothalamic fibers in their study were of fine caliber (,1.5 µm), those emerging from lamina VI exhibited en passant swellings (size #1µm) and tiny varicosities at the tips of short stalks, whereas those that issued from lamina V also bore swellings, but, on average, these were slightly larger (,2 µm), and the majority of them appeared en passant or at the ends of terminal ramifications. However, note that, although their layer V afferent endings were different in size from the type I (layer VI) endings, the difference was slight. In our study of the hamster, we did not specifically differentiate between the laminar origins of cortical afferents to the LP; however, we were unable to group our reconstructed axons according to a strictly dual classification, as reported for the rat (Bourassa and Deschenes, 1995). Moreover, fibers with only en passant swellings and no short appendages, similar to the type II endings described by Bourassa and Deschenes for the rat, are present, but infrequently so, in our material. Finally, in the rat, there appears to be no cortico-LP axon terminal that resembles the medium-sized ending we document for the hamster. Overall, our results show that, instead of the cortical input to the LP being the AFFERENTS TO LATERAL POSTERIOR NUCLEUS ‘‘primary’’ one, it is the tectal projections that have a massive presence throughout much of this nucleus, and these are likely to have the greatest influence in activation of LP neurons [cf. Guillery’s (1995) scheme for connectional relationships of first-order and higher order thalamic nuclei]. Additional comparative studies, including ultrastructural observations, are necessary to resolve these discrepancies in results or even to show definitively that these observations might reflect a species-derived difference in the morphology of cortico-LP projections. SUMMARY AND CONCLUSIONS We have illustrated the morphology of preterminal and terminal axons in the LP, and we have documented a source-specific configuration of the terminal specializations of each afferent system. Although we remain ignorant of the precise significance of these morphological features, it is clear that only by understanding each set of projections at both the light microscopic and the EM level can we hope to uncover structure-function relationships in groups of cells in the central nervous system. For instance, it is likely that the tubular clusters formed by tecto-LP axons represent a highly reliable mode for the transmission of signals between tectal neurons and LP cells (cf. Rouiller et al., 1986), whereas the spiny fibers, which derive from the cortex and which bear tiny varicosities at the tips of their multiple appendages, form an anatomical substrate for a very different type of signaling (Guillery, 1995). The overlapping distribution within the target, together with the unique terminal configurations of afferent endings in the LP, could indicate a predominant influence of the presynaptic neuron in controlling the terminal morphology of axons from different sources. However, such a conclusion would be somewhat premature at this time, because we do not know the detailed relationships between afferent endings and specific subtypes of postsynaptic neurons that may coexist in discrete regions of the LP, with each type of postsynaptic neuron receiving afferents from only one source (however, see Crain and Hall, 1980a). The possibility that specific dendritic domains of the postsynaptic cell can differentially regulate afferent terminal morphology also remains viable. ACKNOWLEDGMENTS Thanks to Jason Glanz for help with the histology and Chrysty Remillard for photographic assistance. We are grateful to Dr. Ray Guillery for his helpful comments on an earlier version of this paper. This work was supported by NIH grants EY 05504 (S.J.), EY 00126 (G.E.S.), and EY 02621 (vision core grant). LITERATURE CITED Abplanalp, P. (1970) Some subcortical connections of the visual system in tree shrews and squirrels. Brain Behav. Evol. 3:155–168. Aggoun-Zouaoui, D., and G.M. Innocenti (1994) Juvenile visual callosal axons in kitten display origin-and fat-reltaed morphology and distribution of arbors. Eur. J. Neurosci. 6:1846–1863. Altman, J., and M.B. Carpenter (1961) Fiber projections of the superior colliculus in the cat. J. Comp. Neurol. 116:157–177. Angelucci, A., F. Clasca, and M. Sur (1996) Anterograde axonal tracing with the subunit B of cholera toxin: A highly sensitive immunohistochemical protocol for revealing fine axonal morphology in adult and neonatal brains. J. Neurosci. Methods 65:101–112. 481 Arango, V., and F. Scalia (1978) Differential projections of the nuclei of the pretectal complex in the rat. Anat. Rec. 190:327. Bennett-Clarke, C.A., R.D. Mooney, N.L. Chiaia, and R.W. Rhoades (1991) Serotonin immunoreactive neurons are present in the superficial layers of the hamster’s, but not the rat’s, superior colliculus. Exp. Brain Res. 85:587–597. Benevento, L.A., and F.F. Ebner (1970) Pretectal, tectal, retina and cortical projections to thalamic nuclei of the opossum in stereotaxic coordinates. Brain. Res. 18:171–175. Benevento, L.A., and J.H. Fallon (1975) The ascending projections of the superior colliculus in the rhesus monkey (Macaca mulatta). J. Comp. Neurol. 160:339–362. Bhide, P.G., and D.O. Frost (1991) Stages of growth of hamster retinofugal axons: Implications for developing axonal pathways with multiple targets. J. Neurosci. 11:485–504. Bourassa, J., and M. Deschenes (1995) Corticothalamic projections from the primary visual cortex in rats: A single fiber study using biocytin as an anterograde tracer. Neuroscience 66:253–263. Bourassa, J., D. Pinault, and M. Deschenes (1994) Corticothalamic projections from the cortical barrel field to the somatosensory thalamus in rats: A single fiber study using biocytin as an anterograde tracer. Eur. J. Neurosci. 7:19–30. Carlson, J., and L. Heimer (1986) The projection from the parataenial thalamic nucleus as demonstrated by the Phaseolus vulgaris-leucoagglutinin (PHA-L) method, identifies a subterritorial organization of the ventral striatum. Brain Res. 374:375–379. Casagrande, V.A., J.K. Harting, W.C. Hall, I.T. Diamond, and G.F. Martin (1972) Superior colliculus of tree shrew: A structural and functional subdivision into superficial and deep layers. Science 177:444–447. Caviness, V.S., Jr. (1975) Architectonic map of neocortex of the normal mouse. J. Comp. Neurol. 164:247–264. Caviness, V.S., Jr., and D.O. Frost (1980) Tangential organization of thalamic projections to the neocortex in the mouse. J. Comp. Neurol. 194:355–367. Clarke, S., F. de Ribaupierre, E.M. Rouiller, and Y. de Ribaupierre (1993) Several neuronal and axonal types form long intrinsic connections in the cat primary auditory cortical field (AI). Anat. Embryol. 188:117– 138. Coleman, J., and W.J. Clerici (1980) Extrastriate projections from thalamus to posterior occipital-temporal cortex in rat. Brain Res. 194:205–209. Coleman, J., and W.J. Clerici (1981) Organization of thalamic projections to visual cortex in opossum. Brain Behav. Evol. 18:41–59. Crain, B.J., and W.C. Hall (1980a) The normal organization of the lateral posterior nucleus of the golden hamster. J. Comp. Neurol. 193:350–370. Crain, B.J., and W.C. Hall (1980b) The organization of the lateral posterior nucleus in neonatal golden hamsters. J. Comp. Neurol. 193:371–382. Crain, B.J., and W.C. Hall (1980c) The organization of the lateral posterior nucleus of the golden hamster after neonatal superior colliculus lesions. J. Comp. Neurol. 193:383–401. Crain, B.J., and W.C. Hall (1980d) The organization of afferents to the lateral posterior nucleus in the golden hamster after different combinations of neonatal lesions. J. Comp. Neurol. 193:403–412. de Venecia, R.K., and N.T. McMullen (1994) Single thalamocortical axons diverge to multiple patches in neonatal auditory cortex. Dev. Brain Res. 81:135–142. Diamond, I.T. (1973) The evolution of the tecto-pulvinar system in mammals: Structural and behavioral studies in the visual system. Sym. Zool. Soc. London 33:205–233. Dursteler, M.R., C. Blakemore, and L.U.J. Garey (1979) Projections to the visual cortex in the golden hamster. J. Comp. Neurol. 183:185–204. Erzurumlu, R.S., S. Jhaveri, and G.E. Schneider (1988) Distribution of morphological different retinal axon terminals in the hamster dorsal lateral geniculate nucleus. Brain Res. 461:175–181. Famiglietti, E.V., and A. Peters (1972) The synaptic glomerulus and the intrinsic neuron in the dorsal lateral geniculate nucleus of the cat. J. Comp. Neurol. 144:285–334. Gerfen, C.R., and P.E. Sawchenko (1984) An anterograde neuroanatomical tracing method that shows the detailed morphology of neurons, their axons and terminals: Immunohistochemical localization of an axonally transported plant lectin, Phaseolus vulgaris-leucoagglutinin (PHA-L). Brain Res. 290:219–238. Gerfen, C.R., and P.E. Sawchenko (1985) A method for anterograde axonal tracing of chemically specified circuits in the central nervous system: combined Phaseolus vulgaris-leucoagglutinin (PHA-L) tract tracing and immunohistochemistry. Brain Res. 343:144–150. 482 Giolli, R.A., L.C. Towns, T.T. Takahashi, A.N. Karamanlidis, and D.D. Williams (1978) An autoradiographic study of the projections of visual cortical area 17 to the thalamus, pretectum and superior colliculus of the rabbit. J. Comp. Neurol. 180:743–752. Glendenning, K.K., J.A. Hall, I.T. Diamond, and W.C. Hall (1975) The pulvinar nucleus of Galago senegalensis. J. Comp. Neurol. 161:419–458. Gonzalez-Hernandez, T.H., D. Galindo-Mireles, A. Castaneyra-Perdomo, and R. Ferres-Torres (1991) Divergent projections of projecting neurons of the inferior colliculus to the medial geniculate body and the contralateral inferior colliculus in the rat. Hearing Res. 52:17–22. Gould, H.J., III, W.C. Hall, and F.F. Ebner (1978) Connections of the visual cortex in the hedgehog (Paraechinus hypomelas). I. Thalamocortical projections. J. Comp. Neurol. 177:445–472. Graybiel, A.M. (1972) Some ascending connections of the pulvinar and nucleus lateralis posterior of the thalamus in the cat. Brain Res. 44:99–125. Graybiel, A.M., and D.M. Berson (1980) Histochemical identification and afferent connections of subdivisions in the lateralis posterior-pulvinar complex and related thalamic nuclei in the cat. Neuroscience 5:1175– 1238. Graybiel, A.M., and D.M. Berson (1981) On the relation between transthalamic and transcortical pathways in the visual system. In F.O. Schmitt, F.G. Worden, and F. Dennis (eds): The Organization of the Cerebral Cortex. Cambridge: MIT press, pp. 285–319. Grove, E.A., V.B. Domesick, and W.J.H. Nauta (1986) Light microscopic evidence of striatal input to intrapallidal neurons of cholinergic cell group Ch4 in the rat: A study employing the anterograde tracer Phaseolus vulgarisleucoagglutinin (PHA-L). Brain Res. 367:379–384. Guillery, R.W. (1969) A quantitative study of synaptic inter-connections in the dorsal lateral geniculate nucleus of the cat. Zeitschrift Fur Zellforschung 96:39–48. Guillery, R.W. (1995) Anatomical evidence concerning the role of the thalamus in corticocortical communications: A brief review. J. Anat. 187:583–592. Guillery, R.W., and G.L. Scott (1971) Observations on ynaptic patterns in the dorsal lateral geniculate nucleus of the cat: The C laminae and the perikaryal synapses. Exp. Brain Res. 12:184–203. Hall, W.C., and F.F. Ebner (1970) Parallels in the visual afferent projections of the thalamus in hedgehog (Pareachinus hypomelus) and turtle (Pseudemys scripta). Brain Behav. Evol. 3:135–154. Harting, J.K., I.T. Diamond, and W.C. Hall (1973) Anterograde degeneration study of the cortical projection of the lateral geniculate and pulvinar nuclei in the tree shrew. J. Comp. Neurol. 150:393–440. Hollander, H., J. Tietze, and H. Distel (1979) An autoradiographic study of the subcortical projections of the rabbit striate cortex in the adult and during postnatal development. J. Comp. Neurol. 184:783–794. Huerta, M.F., and J.I.K. Harting (1984) The mammalian superior colliculus: studies of its morphology and connection. In H. Vanegas (ed): Comparative Neurology of the Optic Tectum. New York: Plenum Publishing, pp. 687–773. Jhaveri, S., D.P.M. Northmore, and G.E. Schneider (1985) Morphology of tectal efferents to diencephalon visualized with the aid of anterograde transport of PHA-L. Soc. Neurosci. Abstr. 11:233. Jones, E.G. (1985) The Thalamus. New York: Plenum Press. Karamanlidis, A.N., and R.A. Giolli (1977) Thalamic input to the rabbit visual cortex: Identification and organization using horseradish peroxidase (HRP). Exp. Brain. Res. 29:191–200. Karamanlidis, A.N., R.P. Saigal, R.A. Giolli, O. Mangana, and H. Michaloudi (1979) Visual thalamocortical connections in sheep studied by means of the retrograde transport of horseradish peroxidase. J. Comp. Neurol. 187:245–260. Kita, H., and S.T. Kitai (1987) Efferent projections of the subthalamic nucleus in the rat: Light and electron microscopic analysis with the PHA-L method. J. Comp. Neurol. 260:435–452. Klein, B.G., R.D. Mooney, S.E. Fish, and R.W. Rhoades (1986) The structural and functional characteristics of striate cortical neurons that innervate the superior colliculus and lateral posterior nucleus in hamster. Neuroscience 17:57–78. LeDoux, J.E., D.A. Ruggiero, and D. Reis (1985) Projections to the subcortical forebrain from anatomically defined regions of the medial geniculate body in the rat. J. Comp. Neurol. 242:182–213. Lee, C.L., D.J. McFarland, and J.R. Wolpaw (1988) Retrograde transport of the lectin Phaseolus vulgaris-leucoagglutinin (PHA-L) by rat spinal motor neurons. Neurosci. Lett. 86:133–138. Lent, R. (1982) The organization of subcortical projections of the hamster’s visual cortex. J. Comp. Neurol. 206:227–242. C. LING ET AL. Lewis, J.W., and J.F. Olavarria (1995) Two rules for callosal connectivity in striate cortex of the rat. J. Comp. Neurol. 361:119–137. Linden, R., and C.E. Rocha-Miranda (1981) The pretectal complex in the opossum: Projections from striate cortex and correlation with retinal terminal fields. Brain Res. 207:267–278. Ling, C., S. Jhaveri, and G.E. Schneider (1994) Anterograde transport of cholera toxin subunit B (CT-B) reveals detailed morphology of optic terminals in certain retinorecipient zones of hamsters. Soc. Neurosci. Abstr. 20:771. Ling, C., S. Jhaveri, and G.E. Schneider (1997) Target- as well as source-derived factors direct the morphogenesis of anomalous retinothalamic projections. J. Comp. Neurol. 388:454–466. Lund, R.D., and T.J. Cunningham (1972) Aspects of synaptic and laminar organization of the mammalian lateral geniculate body. Invest. Opthalmol. 164:287–304. Martin, G.F. (1969) Efferent tectal pathways of the opossum (Didelphis virginiana). J. Comp. Neurol. 135:209–224. Mathers, L.H. (1971) Tectal projection to the posterior thalamus of the squirrel monkey. Brain Res. 35:295–298. McDonald, A.J. (1992) Neuroanatomical labeling with biocytin: A view. Neuroreport 3:821–827. Mikkelsen, J.D. (1992) Visualization of efferent retinal projections by immunohistochemical identification of cholera toxin subunit B. Brain Res. Bull. 28:619–623. Mooney, R.D., M.M. Nikoletseas, S.A. Ruiz, and R.W. Rhoades (1988) Receptive field properties and morphological characteristics of the superior collicular neurons that project to the lateral posterior and dorsal lateral geniculate nuclei in the hamster. J. Neurophysiol. 59:1333–1351. Morest, D.K. (1965) Identification of homologous neurons in the posterolateral thalamus of cat and Virginia opossum. Anat. Rec. 151:390. Morest, D.K. (1968) The collaterals system of the medial nucleus of the trapezoid body of the cat, its neuronal architecture and relation to the olivo-cochlear bundle. Brain Res. 9:288–311. Morest, D.K. (1975) Structural organization of the auditory pathway. In D.B. Tower (ed): The Nervous System, Vol. 3: Human Communication and its Disorders. New York: Raven Press, pp. 19–29. Moya, K.L., L.I. Benowitz, and G.E. Schneider (1990) Abnormal retinal projections alter GAP-43 patterns in the diencephalon. Brain Res. 527:259–265. Nauta, W.J.H., and V.M. Bucher (1954) Efferent connections of the striate cortex in the albino rat. J. Comp. Neurol. 100:257–286. Niimi, K., M. Miki, and S. Kawamura (1970) Ascending projections of the superior colliculus in the cat. Okajimas Fol. Anat. Jpn. 47:269–287. Ogren, M.P., and A.E. Hendrickson (1979) The morphology and distribution of striate cortex terminals in the inferior and lateral subdivision of the Maccaca monkey pulvinar. J. Comp. Neurol. 188:179–200. Olavarria, J., and R.C. Van Sluyters (1985) Organization and postnatal development of callosal connections in the visual cortex of the rat. J. Comp. Neurol. 239:1–26. Perry, V.H. (1980) A tectocortical visual pathway in the rat. Neuroscience 5:915–927. Perry, V.H., and A. Cowey (1982) A sensitive period for ganglion cell degeneration and formation of aberrant retinal-fugal connections following tectal lesion in rats. Neuroscience 7:583–594. Raczkowski, D., and I.T. Diamond (1981) Projections from the superior colliculus and the neocortex to the pulvinar nucleus in Galago. J. Comp. Neurol. 200:231–254. Rafols, J.A., and H.A. Matzke (1970) Efferent projections of the superior colliculus in the opossum. J. Comp. Neurol. 138:147–160. Rafols, J.A., and F. Valverde (1973) The structure of the dorsal lateral geniculate nucleus in the mouse. A Golgi and electron microscopic study. J. Comp. Neurol. 150:303–332. Rajakumar, N., K. Elisevich, and B.A. Flumerfelt (1993) Biotinylated dextran: A versatile anterograde and retrograde tracer. Brain Res. 607:47–53. Rhoades, R.W., S.E. Fish, R.D. Mooney, and N.L. Chiaia (1987) Distribution of visual callosal projection neurons in hamsters subjected to transection of the optic radiations on the day of birth. Dev. Brain Res. 32:217–232. Robson, J.A., and W.C. Hall (1977a) The organization of the pulvinar in the grey squirrel (Sciurus carolinensis). I. Cytoarchitecture and connections. J. Comp. Neurol. 177:355–388. Robson, J.A., and W.C. Hall (1977b) The organization of the pulvinar in the grey squirrel (Sciurus carolinensis). II. Synaptic organization and AFFERENTS TO LATERAL POSTERIOR NUCLEUS comparisons with the dorsal lateral geniculate nucleus. J. Comp. Neurol. 177:389–416. Rockland, K.S. (1995) Further evidence for two types of corticopulvinar neurons. Neuroreport 5:1865–1868. Rouiller, E.M., R. Cronin-Schreiber, D.M. Fekete, and D.K. Ryugo (1986) The central projections of intracellularly labeled auditory nerve fibers in cats: An analysis of terminal morphology. J. Comp. Neurol. 249:261– 278. Schneider, G.E. (1969) Two visual systems: Brain mechanisms for localization and discrimination are dissociated by tectal and cortical lesions. Science 163:895–902. Schneider, G.E. (1973) Early lesions of superior colliculus: Factors affecting the formation of abnormal retinal projections. Brain Behav. Evol. 8:73–109. Schneider, G.E. (1975) Two visuomotor systems in the hamster. Neurosci. Res. Prog. Bull. 13:255–257. Schneider, G.E., and S. Jhaveri (1983) Projection of the internal optic tract: Retinal projections which survive superficial thalamic lesions in hamsters. Soc. Neurosci. Abstr. 9:809. Shu, S.Y., and G.M. Peterson (1988) Anterograde and retrograde transport of Phaseolus vulgaris-leucoagglutinin (PHA-L) from the globus pallidus to the striatum of the rat. J. Neurosci. Methods 25:175–180. 483 Sugita, S., K. Otani, A. Tokunaga, and K. Terasawa (1983) Laminar origin of the tecto-thalamic projections in the albino rat. Neurosci. Lett. 43:143–147. Szentagothai, J. (1963) The structure of the synapse in the lateral geniculate body. Acta Anat. 55:166–185. Szentagothai, J., J. Hamori, and T. Tombol (1966) Degeneration and electron microscope analysis of the synaptic glomeruli in the lateral geniculate body. Exp. Brain Res. 2:283–301. Tamamaki, N., D.J. Uhlrich, and S.M. Sherman (1995) Morphology of physiologically identified retinal X and Y axons in the cat’s thalamus and midbrain as revealed by intraaxonal injection of biocytin. J. Comp. Neurol. 354:583–607. Tarlov, E.C., and R.Y. Moore (1966) The tecto-thalamic connections in the brain of the rabbit. J. Comp. Neurol. 126:403–421. Tiao, Y.-C., and C. Blakemore (1976a) Functional organization in the visual cortex of the golden hamster. J. Comp. Neurol. 168:459–481. Tiao, Y.-C., and C. Blakemore (1976b) Functional organization in the superior colliculus of the golden hamster. J. Comp. Neurol. 168:483– 503. Updyke, B.V. (1975) The patterns of projection of cortical areas 17, 18, and 19 onto the laminae of the dorsal lateral geniculate nucleus in the cat. J. Comp. Neurol. 163:377–395.