Topographic Organization of the Projections from the Entorhinal Area to the Hippocampal Formation of the Rat OSWALD STEWARD D e p a r t m e n t s of Neurological Surgery a n d Physiology, University of V i r g i n i a School of Medicine, Charlottesuille, V i r g i n i a 22901

ABSTRACT The present study re-examines, with autoradiographic methods, the pattern of termination of fibers originating from various medio-lateral divisions of the entorhinal cortex on dentate granule cells and on hippocampal pyramidal cells of the rat. Entorhinal fibers were found to distribute in a proximodistal gradient along the dendrites of dentate granule cells, with afferents from the medial entorhinal area terminating in the innermost portion of the entorhinal synaptic field, afferents from the lateral entorhinal area terminating in the most superficial portions of the entorhinal synaptic field, and intermediate medio-lateral locations in the entorhinal area terminating in intermediate locations in the entorhinal synaptic zone. A similar graded pattern of termination of medial and lateral entorhinal fibers was apparent in the very slight crossed projection of the entorhinal area to the contralateral dentate gyrus. In addition, a comparable gradient in the pattern of termination of entorhinal fibers was evident in the entorhinal projection field in the distal dendritic regions of the pyramidal cells of regio inferior of the hippocampus proper. Entorhinal projections to regio superior were, however, organized in quite a different fashion. In this zone, there was no evidence of a proximo-distal gradient in the patterns of termination of medial and lateral entorhinal areas along the dendrites of regio superior pyramidal cells. Rather, the medio-lateral organization was in a longitudinal dimension, with medial entorhinal afferents terminating in the portions of regio superior near the CAI-CAB transition, and lateral entorhinal afferents terminating furthest from the CA1-CAB transition, immediately adjacent to the CAl-subicular transition, and in the molecular layer of the subiculum proper. A comparable longitudinal organization of entorhinal projections to regio superior was also evident in the zones of termination of the crossed temporo-ammonic tract, contralateral to the injection. These results demonstrate a heretofore unrecognized complexity in the patterns of projection of the entorhinal area to the hippocampal formation, and illustrate that the entorhinal cortex cannot be divided into only two discrete divisions on the basis of the pattern of projection.

The entorhinal area of the rat, which gives rise to the major extrinsic afferent pathway to the hippocamp a1 form ation, has classically been subdivided into a pars medialis and a pars lateralis according to cytoarchitectonic criteria (Krieg, '46; Lorente de NO, '34; and Blackstad, '56), and on the basis of the patterns of efferent projection (Hjorth-Simonsen and Jeune, '72; and Hjorth-Simonsen, '72). The elegant investigations of the latter authors clearly demonstrate that the entorhinal afJ. COMP. NEUR.,167: 285-314

ferents distribute in a selective fashion along the dendrites of the dentate granule cells and hippocampal pyramidal cells of regio inferior, according to their point of origin in the entorhinal area. Several aspects of this entorhinal projection system to the hippocampal formation still remain unclear, however. First, while the projections from the entorhinal area are distributed in a laminated fashion to the dentate gyrus and regio inferior, it is not clear if this distribution reflects the existence of 285

286

OSWALD STEWARD

two discrete and non-overlapping pathways from pars lateralis and medialis, as suggested by histochemical observations (Hjorth-Simonsen and Jeune, '72) or rather whether there is a gradient in a medio-lateral dimension in the mode of termination of entorhinal afferents. HjorthSimonsen ('72) points out that there is a gradual cytoarchitectonic transition between pars medialis and pars lateralis, and suggests that there may be a similar transition in the mode of termination of fibers from the two regions. In fact, HjorthSimonsen ('72) describes one case with a lesion in this transition zone, and demonstrates that degeneration products are found in an intermediate location in the stratum moleculare of the dentate gyrus. The present study extends the observations of Hjorth-Simonsen and Jeune ('72) and Hjorth-Simonsen ('72) and demonstrates that the entorhinal area projects either in a medio-lateral gradient or in a discrete fashion with multiple and probably overlapping subsystems, rather than as two discrete pathways. A second unresolved question involves the very slight normal crossed pathway from the entorhinal area to the contralateral dentate gyrus, which has recently been described (Zimmer and Hjorth-Simonsen, '75; Goldowitz et al., '75). This normally light crossed projection proliferates extensively following ipsilateral entorhinal lesions (Steward et al., '73, '74a; Zimmer and Hjorth-Simonsen, '75), and in the final post-lesion state assumes a laminated organization which is highly reminiscent of the pattern of termination of medial and lateral entorhinal fibers in the ipsilateral dentate gyrus (Steward et al., '74b; Zimmer and Hjorth-Simonsen, '75). An understanding of the possible mechanisms of this specificity of termination of the lesion induced crossed projections must rest on a clear description of the pattern of termination of the normal crossed system. In the present study, the slight normal crossed projections are also shown to be organized in a laminated fashion, according to their site of origin in the entorhinal area. Finally, fibers originating in the entorhinal area also project to the distal dendritic regions of the pyramidal cells of regio superior, both ipsilaterally and contralaterally (Blackstad, '56; Raisman et al.,

'65; Steward et al., '73, '74a). However, the topographic organization of this projection system is still in doubt. Zimmer and Hjorth-Simonsen ('75) report that the major contributor to the crossed temporoammonic tract is the medial entorhinal area, but that lesions which involved the lateral entorhinal area resulted in equivocal degeneration debris at the transition between CA1 and subiculum contralaterally. Several studies of ipsilateral projections, however, have demonstrated that lesions which involve the lateral entorhinal area and peri-amygdaloid cortex resulted in degeneration products in the CAl-subicular transition zone, but not in the CA1 dendritic zones further removed from the subiculum (Cragg, '61; Powell et al., '65). The present study analyzes the entorhinal projections to regio superior both ipsilaterally and contralaterally, and demonstrates that the medial entorhinal area gives rise to projections to portions of the CA1 region near the CA1-CA2 transition both ipsilaterally and contralaterally, and in addition, projects to all septo-temporal levels of the hippocampal formation. The lateral entorhinal area, on the other hand, projects to the portion of CA1 furthest from the CA1-CAB transition (closest to the subiculum in the ventral hippocampus). While this projection is bilateral to the rostra1 pole of the hippocampus, fibers from the lateral entorhinal region apparently fail to reach more temporal sites in the contralateral hippoc ampal form ation. MATERIALS AND METHODS

The projections of the various divisions of the entorhinal area were traced autoradiographically in a total of over 40 adult male Sprague-Dawley rats. The entorhinal region was approached stereotaxically from the dorsal aspect, at a n angle of 10" from the midline. With the aid of a Hamilton micro syringe, 10-15 ,Ci (in 1 ,I) of 3H proline or leucine (New England Nuclear, specific activity = 30-50 Ci/mM) was delivered over a period of 45 minutes. Three to six days post-injection, the animals were deeply anaesthetized with sodium pentobarbital, and perfused via a trans-cardial route with 10% formalin. The brains were removed, post-fixed in 10% formalin for three days, and subsequently processed for autoradiography

TOPOGRAPHY OF ENTORHINAL PATHWAYS

either by paraffin embedding according to methods previously described (Steward et al., '74), or by embedding in egg yolk according to the method of Ebbesson ('70). The brains were divided with a coronal section at approximately the posterior border of the dorsal psalterium, and the rostra1 portion of the brain was sectioned in the coronal plane, while the caudal region was sectioned in the horizontal plane. This combined coronal/ horizontal sectioning technique was chosen since the hippocampus curves from the septum back in a caudal and ventral direction. Thus, any single sectioning plane does not permit a clear analysis of the various subfields of the hippocampal formation, and in particular, does not provide a section which is approximately parallel to the long axis of the cells constituting the hippocampal formation. Sections parallel to the long axis of the granule and pyramidal cells were particularly important in analyzing the distribution of entorhinal afferents along the dendritic zones. The autoradiographic method was essentially comparable to that described by Cowan et al., '72. The egg yolk embedded brains were sectioned at 20-30 pm on a freezing microtome, while the paraffin embedded brains were sectioned at 10 /*m. Both the frozen and paraffin sections were mounted on slides with gelatin, defatted through xylene, and rehydrated to water, and finally coated with Kodak NTB-2 emulsion. After a 25-30 day exposure, the autoradiographic preparations were developed with Kodak D-19, stained with Cresylecht violet, and covered. There were no qualitative or quantitative differences between the preparations which were prepared for paraffin embedding, and those which were sectioned on the freezing microtome, and thus no distinction is made between the cases on this basis. Entorhinal projections were also analyzed in silver stained preparations following entorhinal lesions. Twenty-four hours to five days following the lesions, the animals were deeply anaesthetized with sodium pentobarbital, and perfused transcardially with 10% formalin. The brains were subsequently sectioned on the freezing microtome, and processed according to a slightly modified Fink-Heimer ('67) procedure.

287

OBSERVATIONS

Subdivisions of the entorhinal region Prior to an investigation of the topographic organization of entorhinal cortical afferents, a description of the cytoarchitectonic subdivisions of the entorhinal region itself is perhaps in order. The entorhinal area (Broadman's area 28) has been variously subdivided on the basis of a variety of criteria, and in a variety of species (Lorento de NO, '34; Cajal, '11; Krieg, '46; Van Ferreira, '51; Blackstad, '56). The parcellation of the entorhinal region adopted in the present paper is for the most part consistent with that of Blackstad ('56). Blackstad ('56) utilized silver impregnation, and cell stains for his parcellation of the entorhinal region, and divided the region on the basis of both cyto- and fibroarchitectonics. Since the autoradiographic material was stained by Cresyl violet, however, the division of the entorhinal area in the present report is based exclusively on Cresyl violet stained sections. While the criteria for the subdivision differ slightly, the areas defined are not substantially different from those of Blackstad ('56). The total extent of the entorhinal region is defined as extending from the border with the parasubiculum to the rhinal fissure in ventral sections, and in dorsal segments, where the rhinal fissure is not prominent to the border with the adjacent neocortex. Within this region, several subareas can be recognized. The most medial portion immediately adjacent to the parasubiculum has been called area entorhinalis pars medialis (aepm) (Blackstad, '56). This zone is characterized in Cresyl violet stained sections by a prominent layer 11, consisting of tightly packed stellate cells (Lorente de NO, '34; Blackstad, '56). Layer 111, however, which is comprised largely of medium sized pyramidal cells (Lorente de NO, '34) is relatively cell poor, in comparison to the other subareas of the entorhinal region. While other criteria can be used to differentiate this area, particularly the presence of the lamina dissecans, and the appearance of layers V-VI, these criteria are not utilized in the present analysis, since we show in another communication that it is the cells of layers I1

288

OSWALD STEWARD

and I11 which give rise to the afferents to the hippocampal formation (Steward and Scoville, '76). Moving laterally, a subarea of the entorhinal region can be recognized which is intermediate between pars medialis and pars lateralis. While this transition zone was describd by Blackstad ('56) and by Lorente de NO ('34) the dorso-ventral organization was not analyzed. This transition zone is characterized in Cresyl violet stained sections by a less compact layer I1 than in pars medialis, comprised in part of somewhat smaller fusiform cells (Blackstad, '56). The cell density in layer I11 seems greater than in pars medialis. In addition, these characteristics are further accentuated in pars lateralis. In the rat, this transition zone does not exhibit the cell islands in layer I1 which characterize pars lateralis, but there is a tendency for a discontinuity of layer I1 which suggests cell groupings. This transition zone is illustrated in figures 1 and 2, and is designated area entorhinalis pars intermedialis (aepi). Still more laterally in the entorhinal region, is the zone which has classically been recognized as area entorhinalis pars lateralis (aepl, figures 1 and 2 ) . This zone is characterized primarily by the cell islands i n layer 11, and by the more fusiform shape of cells of this lamina. In addition, layer I1 divides into two more or less distinct cell lamina with the cell islands characterizing the more superficial of these laminae (Blackstad, '56). Pars lateralis, as here defined, does not differ substantially from the region similarly designated by Blackstad ('56). At the lateral edge of pars lateralis, particularly in ventral sections, lies a transition zone which seemingly has not been previously described (see figs. 1, 2E-H). This zone does not appear to be identical with the perirhinal area (Broadman's area 35) differing from the perirhinal area in that the cells are more scattered, somewhat larger, and stain less intensely with Cresyl violet than the granule cells of the perirhinal area. While the perirhinal cortex forms a transition cortex between the entorhinal area and neocortex with some cytoarchitectonic characteristics of each (Krieg, '46), the zone under consideration shares all the characteristics of aepl except that the cell islands of layer I1 cease. A

characteristic layer I11 continues for some distance into the depth of the rhinal sulcus, until gradually merging into the perirhinal area. The tendency for increased cell packing in layer 111 is more evident here than in pars intermedialis or lateralis and the layer becomes thinner. This transition zone is only present in ventral segments of the entorhinal area (figs. 2E-H), and is particularly apparent at the transition from the entorhinal cortex to the prepyriform cortex (Price and Powell, '71, their figure 6 ) . It may be that this zone characterizes the transition from area entorhinalis to the prepyriform cortex, which is at more ventral levels insinuated between aepl and the perirhinal area. The perirhinal area, on the other hand borders the rhinal sulcus throughout its extent, and forms the transition between neocortex and paleocortical regions in general (Krieg, '46). This zone has been labeled PRh in figure 2 , and may correspond to the prorhinal area (particularly Pr-1 ) described recently in the monkey (Van Hoesen and Pandya, '75a). As indicated by figures 1-2, the parcellation of the entorhinal area must be considered in a dorso-ventral dimension, as well a s a medio-lateral. For this purpose the medio-lateral extent of the various subdivisions of the entorhinal area was analyzed in the dorso-ventral series of sections illustrated in figures 1-2. The dorso-ventral levels illustrated are separated by approximately 700 pm, and include almost the entire dorso-ventral extent of the entorhinal region as here defined. From these sections, the entorhinal area as reconstructed in unfolded form is illustrated in figure 3, as it would appear from the posterior aspect of the brain. This diagram of the entorhinal region differs little from that of Blackstad ('56) except that the area entorhinalis pars intermedialis is included in the analysis, along with area PRh. As is evidenced from the reconstruction, the most dorsal part of the entorhinal area is almost exclusively comprised of a cytoarchitectonic field meeting the definitions of pars medialis. Moving ventrally, pars intermedialis is next to appear, and its dorso-ventral extent is almost as great as pars medialis. However, clearly differentiated pars lateralis does not appear until level C (fig. 2E), at a level where the

TOPOGRAPHY OF ENTORHINAL PATHWAYS

289

290

OSWALD STEWARD

TOPOGRAPHY OF ENTORHINAL PATHWAYS

291

Fig. 3 Dorso-ventral analysis of the various cytoarchitectonic divisions of the entorhinal area. The reconstruction of the present figure represents the appearance of the entorhinal area as it would appear from a position posterior to the caudal pole of the brain. The levels indicated represent the dorso-ventral levels illustrated in figures 1 and 2.

rhinal fissure becomes well defined. These differencesin the dorso-ventral locations of the various subdivisions of the entorhinal region are important for the analyses of the present paper, since at dorsal locations, the lateral-most part of the entorhinal area is not pars lateralis as classically defined, but rather pars intermedialis. Thus, the cytoarchitectonic medio-lateral axis of the entorhinal area is tilted such that the lateral portion at the entorhinal area is found ventral to the medial entorhinal area.

Entorhinal terminal fields in the hippocampal formation The complete pattern of termination of entorhinal fibers, and the major areal and cytoarchitectonic divisions of the hippocampal formation are summarized in figure 4 (from silver stained preparations). The most massive projection of the ento-

rhinal area is to the ipsilateral dentate gyrus, as noted by previous authors (Blackstad, '56; Raisman et al., '65; HjorthSimonsen and Jeune, '72; and HjorthSimonsen, '72). Silver stains following complete entorhinal lesions (which include the entire medio-lateral extent of the entorhinal area) result in dense degeneration debris in the outer two-thirds of the stratum moleculare of the dentate gyrus, which contains the distal dendritic ramifications of the granule cells. Where the stratum moleculare of the fascia dentata merges with the stratum moleculare of regio inferior of the hippocampus proper, the terminal field in the dentate is continuous with a bulb shaped terminal field in stratum lacunosum-moleculare of regio inferior which contains the most distal dendrites of the pyramidal cells of regio inferior, particularly subzones CA2 and CA3. This projection to the fascia dentata and to

292

OSWALD STEWARD

, Fig. 4 Schematic representation of the hippocampal formation in coronal (upper) and horizontal (lower) sections following unilateral entorhinal lesions (extent of a typical lesion is shown by the shaded area on the left). The pattern of degeneration i n the hippocampus and area dentata is illustrated by the stippling. RS, regio superior; RI, regio inferior; AD, area dentata; Sub, subiculum; Presub, presubiculum; Parasub, parasubiculum; AEPM, area entorhinalis pars medialis; AEPL, area entorhinalis pars lateralis; RF, rhinal fissure; SLM, stratum lacunosum-moleculare; SM, stratum moleculare (of area dentata); SG, stratum granulare.

regio inferior is largely restricted to the ipsilateral hippocampal formation, although recent evidence indicates the existence in the rat of a very slight crossed projection to the contralateral dentate gyrus (zimmer and Hjorth-Simonsen, '75; and Goldo-

witz et al., '75). The slight normal crossed projection can only be observed in silver stained preparations at relatively short post-lesion survival times (24-72 hours, see Goldowitz et al., '75) but can be observed i n autoradiographic preparations.

TOPOGRAPHY OF ENTORHINAL PATHWAYS

A crossed projection to the contralateral CA2-CA3 dendritic zone has not been reported in silver stained material, but some autoradiographic evidence for a very slight contralateral labeling of this zone in regio inferior has been reported by Goldowitz et al. ('75). In the experimental material of the present study, no evidence of such a normal crossed projection to the contralateral regio inferior has been observed. The only truly bilateral entorhinal projection system (in the sense of having approximately equal density on both sides) is the projection to the stratum lacunosummoleculare of regio superior' which contains the most distal tips of pyramidal cell dendrites. The crossed pathway to this zone was initially described by Cajal ( ' l l ) , and termed the crossed temporo-ammonic tract. Both the ipsilateral and crossed pathways have recently been demonstrated experimentally (Steward et al., '73, '74a). Silver stained preparations indicate that the ipsilateral projections to regio superior extend at least from the stratum moleculare of the subiculum, near the border with the CA1 region, to the CA1-CA2 transition zone (fig. 4 ) . Contralaterally, a similar organization may be observed in the rostra1 hippocampal formation. Further caudally, however, the density of the crossed temporo-ammonic projection decreases (on the basis of silver stains and autoradiography ) . In addition, the crossed temporoammonic projection changes its distribution slightly, since degeneration products are found only in the stratum lacunosummoleculare of the CA1 region, while the stratum moleculare of the subiculum is free of degeneration debris. While the large lesions utilized to trace these entorhinal projection systems also involve portions of the parasubiculum, presubiculum, and occasionally the subiculum proper, autoradiographic evidence (personal observations) and the retrograde labeling of the cells of origin of entorhinal afferents (Steward and Scoville, '76) suggest that the entorhinal afferents to the hippocampal formation arise exclusively from the entorhinal region proper, from cells in layers I1 and 111. Topography of temporo-dentate projections In order to investigate the topographic organization of the entorhinal projections

293

to the dentate gyrus, photographs were taken from a standard site in the dorsal leaf of the dentate as viewed in coronal section. The zone photographically analyzed extended from the stratum granulosum (SG of figs. 5-8) up beyond the distal tips of the granule cell dendrites into the stratum lacunosum-moleculare of regio superior. For the analysis of the projections to the dentate gyrus, 10 exemplary cases were chosen from a total of over 40 brains with injections in the entorhinal area. Some of the animals chosen for photographic illustration had sustained unilateral entorhinal lesions opposite to the injection for the purpose of tracing lesion induced crossed projections. The ipsilatera1 temporo-dentate system in these animals was entirely comparable to unoperated animals, and these cases were thus utilized for the purpose of illustrating the normal pattern of ipsilateral projection. The exemplary cases are divided into four groups, according to the location of the injection along the medio-lateral axis of the entorhinal cortex. The cases with injections localized to the most medial portion of the entorhinal area are illustrated in figure 5. In the first of these cases (MER/P-1, fig. 5C), the injection is centered over the pre- and parasubiculum, and spreads to the most medial portion o f layers 2-3 of the entorhinal area proper. The injection in the second case (MER/P-2, fig. 5D) is localized to layers 2-3 of the medial entorhinal area proper. The center of the injection site lies much deeper in the entorhinal area, however, than in case MER/P-1. Figures 5A,B illustrate the stratum moleculare of the dentate gyrus in these two cases. In both cases, silver grains are confined to the middle one-third of the stratum moleculare, approximately midway out the granule cell dendrites. The more distal dendritic regions (from the labeled terminal field out to the hippocampal fissure) are relatively free of label. In addition, a proximo-distal gradient of grain density is apparent within the middle one-third of the stratum moleculare. The greatest grain density is 1 It is probable that the terminal field of the ipsilateral and crossed temporo-ammonic tracts occupies only the stratum moleculare and does not spread significantly into the stratum lacunosum. The clear distinction between these two laminae is difficult in the rat, however, (except i n reduced silver preparations), and the two laminae have, traditionally been considered together (Blackstad, 5 6 ) . This convention will be followed i n the present paper.

294

OSWALD STEWARD

Fig. 5 Stratum moleculare of the dentate gyrus (photos extend from the stratum granulare ( S G ) up through the stratum lacunosum-moleculare of regio superior). The end of the stratum moleculare is marked by the hippocampal fissure (HF). (A) animal MER/P-1. The injection site for MER/P-1 is shown in C. ( B ) animal MER/P-2. The injection site for MER/P-2 is shown in D. Light field photographs are combined with dark field photographs of the same site i n figures 5, 7, 9, and 11 in order to correlate the bands of grains with the cell body layers. The arrows in C and D illustrate the lateral most boundaries of the entorhinal area. The calibration bar in D represents 50 pm for A and B and 500 p m for C and D.

TOPOGRAPHY O F ENTORHINAL PATHWAYS

found in the entorhinal zone most proximal to the stratum granulosum, and grain density decreases further distally, even within the middle one-third of the stratum moleculare. A similar pattern of labeling in the dentate stratum moleculare was obtained following all injections which labeled the most medial portion of pars medialis, whether these injections were centered in the most dorsal portions of the entorhinal region, or further ventrally (see fig. 6 for a diagrammatic summary of all injection sites which yielded this pattern of termination). Injections which label the medial entorhinal area, but which are not immediately adjacent to the parasubiculum are illustrated in figure 7. In the first of these (MER/'C-1, fig. 7C), the label is quite heavy, and the injection site is wedge shaped with superficial zones being the most heavily labeled. The injection in the second case (MER/C-2, fig. 7 D ) , is considerably wider, and again heavily labels the superficial zones, including layers 2 and 3. As indicated by figure 8 , these injections probably label the most dorsal portion of aepi at the lateral limit of the entorhinal area. In these two examples, grains are found throughout the middle one-third of the stratum moleculare of the dentate gyrus. In contrast to the two cases illustrated in figure 5, there is no indication of a proximo-distal gradient of grain density within the middle stratum moleculare. The dorso-ventral position of these injections is illustrated in figure 8. In the cases illustrated in figure 9 (MLER-7, MLER-8), the injections are centered in the transition zone between medial and lateral entorhinal areas (aepi), at a relatively dorsal site. Note that while these injections are at the lateral limit of the entorhinal area, they lie within pars intermedialis rather than pars lateralis due to their dorsal location (fig. 10). These two injections are virtually identical in terms of their position along the medio-lateral axis of the entorhinal area, and as illustrated in figures 9A,B, their terminal fields in the dentate stratum moleculare are quite comparable. I n these cases, the silver grains are concentrated slightly more distally in the stratum moleculare than is the case in the examples of figures

295

Fig. 6 The dorsol-ventral location of the injection sites of figures 5C, D are illustrated as they would appear from a posterior view of the entorhinal cortical region. The dorso-ventral location of the center of the injection site is indicated by the horizontal line for animals MER/P-1, and MER/P-2. These represent the dorso-ventral location of the horizontal sections which are illustrated in figures 5C, D. MER-3 illustrates the dorso-ventral location of the injection site illustrated in figure 13E.

5 and 7. In fact, the center of the labeled lamina seems to lie approximately at the transition between the middle one-third and the outer one-third of the stratum moleculare, leaving a small portion of the total entorhinal terminal field relatively free of label at both proximal and distal extremes. These two cases of figure 9 are approximately comparable to case S i l l 9 of Hjorth-Simonsen ('72) which also demonstrates an intermediate pattern of projection to the stratum moleculare of the dentate gyrus. As illustrated by figure 10, only two cases are available at present which are centered in and relatively selective to pars intermedialis. In the final two cases, the injections are localized to the lateral and rostral most portions of the entorhinal area, immediately subjacent to the rhinal fissure. In addition, these are somewhat more ventral in

296

OSWALD STEWARD

Fig. 7 S t r a t u m moleculare of t h e d e n t a t e gyrus: ( A ) a n i m a l M E R / C - I . T h e injection s i t e for MER/C-1 i s illustrated in C. ( B ) a n i m a l MER/C-2. T h e injection s i t e for MER/C-2 i s illustrated in D . Abbreviations a n d magnification a r e a a r e a s for figure 5 .

the entorhinal region than are the preceding cases. In both cases (LER-1, LER-2, fig. l l ) , the center of the injection site actually lies on the neocortical side of the

rhinal fissure. In both cases, however, it appears that a significant spread of label occurs into the entorhinal area, and the injection probably labels the most lateral

TOPOGRAPHY OF ENTORHINAL PATHWAYS

297

portion of the dentate stratum moleculare originate from the lateral entorhinal area proper, and probably area PRh. In figures 5-12, the temporo-dentate projections are illustrated in coronal sections. The justification for this choice is that the slight normal crossed temporo-dentate projection is restricted to the most rostral tip of the dentate gyrus, which can only be viewed in coronal section, thus facilitating comparisons between ipsilateral and crossed systems. However, a comparable organization of entorhinal afferents to the ipsilateral dentate gyrus is evident more caudally, in that portion of the dentate gyrus which is visible in horizontal section.

Fig. 8 The dorso-ventral location of the injection sites of figures 7C,D are illustrated as they would appear from a posterior view of the entorhinal cortical region. The dorso-ventral location of the center of the injection site is indicated by the horizontal line for animals MER/C-1 and MER/C-2. These represent the dorso-ventral location of the horizontal sections which are illustrated in figures 7C,D.

portion of pars lateralis, area PRh, the perirhinal area, and adjacent temporal neocortex. As illustrated in figs. 11A,B both of these injections result in a labeled lamina in the outer most stratum moleculare of the dentate gyrus, immediately adjacent to the hippocampal fissure ( H F ) , leaving the middle one-third label free. As illustrated by figure 12, this pattern of labeling was obtained in a total of three cases. With these cases, i t is not possible to exclude the possibility that the afferents to the outermost dentate stratum moleculare actually originate i n the perirhinal area, or even i n the perirhinal neocortex. I n three cases, however, which are not illustrated, the injections labeled only the neocortical area adjacent to the rhinal fissure, and in this case, no projections to any portion of the hippocampal formation were observed. Thus, it seems probable at this time that the afferents to the outermost

Crossed temporo-dentate afferents Considering next the slight normal crossed projection to the contralateral dentate gyrus, recently described by Zimmer and Hjorth-Simonsen ('75), and Goldowitz et al. ('75), figure 13 indicates that medial and lateral entorhinal afferents do distribute to different zones of the stratum moleculare, even on the contralateral side. As reported by Goldowitz et al., the normal crossed projection is heaviest in the ventral blade of the dentate (or the exposed blade according to the terminology of Zimmer and Iljorth-Simonsen, ('75), and this is the zone illustrated in figure 13. Figure 13A illustrates the dentate stratum moleculare ipsilateral to the medial entorhinal injection, diagramatically illustrated in figure 6 (MER-3), and shown in figure 13F. Figure 13B illustrates a comparable zone following the lateral entorhinal injection illustrated in figure 12 (LER-3), and figure 13F. The terminal fields in matched portions of the contralatera1 dentate stratum moleculare are illustrated in figures 13C,D. Because the normal crossed projection is so sparse, i t is difficult to obtain a sufficient amount of label to determine whether there is a gradient in the pattern of projection, but it is clear from figure 13 that the normal crossed temporo-dentate pathway is topographically organized along the granule cell dendrites in a manner comparable to the much more massive ipsilateral temporo-dentate system. Projections t o regio inferior Entorhinal afferents to regio inferior of

298

OSWALD STEWARD

Fig. 9 Stratum moleculare of the dentate gyrus ( A ) animal MLER-7. The injection site for MLER-7 is shown in C. ( B ) animal MLER-8. The injection site for MLER-8 is shown i n D. Abbreviations and magnification are as for figure 5.

the hippocampus proper also distribute in a laminated fashion. In this bulb shaped zone (fig. 14B), the dendrites of the pyramidal cells of regio inferior collect in a manner reminiscent of the hub of a spoked wheel. Along more distal tips of these col-

lected dendrites, entorhinal afferents terminate in the manner illustrated by silver staining in figs. 14A,B. In figures 15A-D, this same zone is illustrated following a series of injections which progress from the most medial (fig. 15A) to the most

TOPOGRAPHY OF ENTORHINAL PATHWAYS

.

299

tion along the length of the dendrites, it was anticipated that a similar organization might characterize the projections to regio superior. This anticipation was not, however, realized. Figure 16 illustrates the zone of entorhinal termination in regio superior as revealed by silver staining following complete entorhinal lesions. Degeneration products may be found in the stratum moleculare of regio superior from the CA1-CAB transition into the portions of the subiculum proper. As figure 17 illustrates, again with a series covering a similar range of injection sites as figures 4-7, the most medial portions of the entorhinal area project to a portion of the stratum lacunosum-moleculare of regio superior furthest from the subiculum (near the CA1, CA2 transition). Progressively more lateral injections result in selective labeling of portions of the stratum lacunosummoleculare more proximal to the subiculum, until with the most laterally placed injection of the series (fig. 17D), the grains are concentrated primarily at the Fig. 10 The dorso-ventral location of the inCAl-subicular transition, and in the strajection sites of figure 8 are illustrated as they would appear from a posterior view of the entotum moleculare of the subiculum proper, rhinal cortical region. The dorso-ventral location leaving zones further from the CAl-subicof the center of the injection site is indicated by ular transition free of label. This organizathe horizontal line for animals MLER-7 and tion is maintained throughout the rostroMLER-8. These represent the dorso-ventral locacaudal axis of the hippocampal formation tion of the horizontal sections which are illustrated in figures 9C,D. ipsilateral to the injection. In rostral segments, where there is no subiculum, it is lateral (fig. 15D). (The injection sites are that portion of CA1 furthest from the CA1, defined in the figure legends). With medial CA2 transition which receives the projecentorhinal injections, the outer rim of the tions from the lateral entorhinal area. The terminal field of the crossed temhub is labeled, whereas with lateral entorhinal injections, the center of the hub is poro-ammonic tract is organized in a simiselectively labeled. Despite the fact that lar manner, at least in more rostra1 the lamination of afferent input is less portions of the hippocampus. In caudal reconspicuous, being concentric rather than gions, however, where the density of the planar owing to the contorted form of the crossed temporo-ammonic projection deneuropil at this dendritic collection site, it creases, and the total terminal field is is clear that a gradient exists in the pat- restricted to the stratum lacunosum-motern of projection of the entorhinal area. leculare of the CA1 region proper, this The organizational feature in regio infe- organization is modified slightly. From the rior, analogous to the projections to the present material, only the medial and trandentate gyrus, is that medial entorhinal sitional zones of the entorhinal cortex seem regions innervate proximal dendritic sites, to give rise to caudally directed crossed while the lateral regions terminate more temporo-ammonic projections, leaving the CAl-subicular transition zone of the caudistally. dal hippocampus free of crossed input. Projections t o regio superior However, within the zones which do reSince the conspicuous organizational ceive the crossed projections, the mediofeature of the entorhinal projections to the lateral organization is exactly comparable fascia dentata and regio inferior is lamina- to the ipsilateral temporo-ammonic system,

300

OSWALD STEWARD

Fig. 11 Stratum moleculare of the dentate gyrus. ( A ) animal LER-1. The injection site for LER-1 i s shown in C. ( B ) animal LERB. T h e injection site for LER-2 is shown i n D. Abbreviations a n d magnification are as for figure 5.

with medial entorhinal cortical regions projecting to the CA1-CA2 transition zone furthest from the subiculum, and lateral regions projecting to portions of the stra-

tum lacunosum -moleculare immediately adjacent to the subiculum. In the present autoradiographic preparations, a clear distinction between a ter-

TOPOGRAPHY OF ENTORHINAL PATHWAYS

30 1

also extends into the more medially located portions of the entorhinal area. In the case illustrated in figure 18A, the zone of terminal degeneration in the dentate stratum moleculare is somewhat thinner than in the case illustrated in figure 18B, owing to the greater medio-lateral extent of the lesion in the case of 18B. In addition, the ipsilateral temporo-ammonic projections to regio superior show a medio-lateral progression from the CA1-CA2 transition zone (fig. 1 8 A ) , back toward the CAl-subicular transition (fig. 1 8 B ) quite comparable to that illustrated autoradiographically i n figure 17. In both these sites in both regio superior, and regio inferior, as well as in the stratum moleculare of the dentate gyrus, the degeneration debris has the appear ance of termin a1 degeneration, rat her than axonal degeneration, suggesting that the various fields described in the autoradiography section do reflect terminal fields rather than pathways for fibers of Fig. 12 T h e dorso-ventral location of the inpassage. A clear demonstration of this disjection sites of figure 11 are illustrated as they tinction must, however, await electron miwould appear from a posterior view of the entorhinal cortical region. The dorso-ventral. location croscopic investigations. of the center of the injection site is indicated by The route of entorhinal fibers into these the horizontal line for animals LER-1 a n d LER-2. terminal fields is also revealed by the These represent the dorso-ventral location of the silver preparations. The route of the fibers horizontal sections which are illustrated in figures into the terminal fields in the fascia 1lC,D. LER-3 illustrates the dorso-ventral location of the injection site illustrated in figure 13F. dentata and regio inferior has been described previously (Blackstad, '56, '58; minal field and fibers of passage is at best Hjorth-Simonsen and Jeune, '72) and will tenuous. Figure 18 illustrates the organiza- not be considered here. However, the route tion of entorhinal projections to regio su- followed by the fibers to regio superior deperior and the internal leaf of the dentate serves some comment. As Hjorth-Simongyrus following both subtotal (figs. 18A,B) sen and Jeune ('72) and Nafstad ('67) and complete (fig. 1 8 C ) entorhinal cor- have shown, fibers from the medial portical lesions. Unfortunately, due to the tions of the entorhinal area en route to inaccessibility of the entorhinal area, le- the portion of CA1 near the CA1-CAB transions, restricted to a single site along the sitional first enter the angular bundle then medio-lateral axis are difficult, and only leave it to cross the stratum moleculare of two sub-total lesions are available at the the subiculum, and stratum lacunosumpresent time. Figure 18A illustrates the moleculare of C A I (figs. 18A,B). In the terminal degeneration in the dentate gyrus case of the lesion of figure 18B, which inand regio superior following a selective volves pars medialis of the entorhinal medial entorhinal lesion, and figure 18B area, some degenerating fibers also travel illustrates a comparable portion of the through the superficial portions of the hippocampal neuropil following a slightly alveus, contributing a large component of larger lesion which was centered in the fibers to the dorsal psalterium (fig. 1 9 B ) . transition zone between pars medialis and Since all three lesions destroy portions of pars lateralis of the entorhinal area. The the presubiculum, while only those involvlesion of figure 18A is roughly comparable ing pars intermedialis, or lateralis result to the injection of figure 7C, while the le- in significant numbers of degenerating sion of figure 18B is roughly comparable fibers in the alveus, it is unlikely that this to the injection illustrated in figure 7C, but alvear component arises entirely from the

302

OSWALD STEWARD

Fig. 13 Comparison between the projections to the ipsilateral and the contralateral dentate stratum moleculare. A-D are paired light field/dark field photographs of the stratum moleculare of the ventral (or exposed) blade of the dentate gyrus. ( A ) ipsilateral to an injection localized to the medial entorhinal area. ( B ) ipsilateral to a n injection localized to the lateral entorhinal area. (C) contralateral to the specimen of A. ( D ) contralateral to the specimen of B. The injection site for A and C is illustrated in E. The dorso-ventral location of this horizontal section is shown i n figure 6 (MER-3). The injection site for B and D is illustrated i n F, while the dorso-ventral location of this injection site is shown i n figure 12 (LER-3).

TOPOGRAPHY O F ENTORHINAL PATHWAYS

303

Fig. 14 Projections of the entorhinal area to regio inferior, as illustrated in coronal silver stained sections. The bulb shaped terminal field encompassing the most distal dendrites of the pyramidal cells of regio inferior is illustrated i n B, from the zone delimited by the rectangle. CAI, C A I sector of the hippocampus; SG, stratum granulare of the fascia dentata; RI, regio inferior of the hippocampus proper; HF, hippocampal fissure. The boundaries between the stratum lacunosum-moleculare of regio inferior and the adjacent fields of CAI and the fascia dentata are marked by the dotted lines.

presubiculum. A most interesting group of fibers can only be observed following lesions which include the lateral entorhinal area, however (figs. 17C, 18C). In this

case, a component of fibers can be seen to percolate down through the stratum pyramidale and moleculare of the subiculum (fig. 1 9 C ) . Upon traversing the stratum

304

OSWALD STEWARD

TOPOGRAPHY OF ENTORHINAL PATHWAYS

305

Fig. 16 Projections of the entorhind area to regio superior, as illustrated i n horizontal silver stained sections. The photograph in B illustrates the region delimited by the rectangle in A. CA1, CAI sector of the hippocampus; SG, stratum granulare of the fascia dentate; RI, regio inferior of the hippocampus; HF, hippocampal fissure; Sub, subiculum.

pyramidale of the subiculum these fibers seem to break up into a sparse terminal field, even before entering the stratum moleculare. The lesion of figures 18C, 19C also encroaches upon the subiculum, and degeneration debris could thus be interpreted as reflecting fibers of passage from points other than the lateral entorhinal

area. However, autoradiographic procedures also suggest the existence of such a projection crossing the subiculum, as illustrated in figure 17. This projection from the lateral entorhinal area to the subiculum corresponds well with the distribution of the so-called alvear path, which was described by Lorente de NO in Golgi mate-

306

OSWALD STEWARD

Fig. 17 Projections of t h e entorhinal area to regio superior, a s illustrated by autoradiographic methods. The zone photographically illustrated is comparable to t h e zone of figure 16B. A-D illustrate the entorhinal terminal field following a series of injections proceeding from the most medial ( A ) to the most lateral ( D ) . T h e injection sites for these cases are as follows: ( A ) MER/P-l (figs. 5A,C); ( B ) MER/C-1 (figs. 7B,D); ( C ) MLER-7 (figs. 9B,D); ( D ) LER-2 (figs. 11B,D).

TOPOGRAPHY OF ENTORHINAL PATHWAYS

Fig. 18 Pattern of degeneration i n the hippocampal formation following partial ( A and

B), and complete ( C ) entorhinal cortical lesions. Lesions are described in the text.

307

308

OSWALD STEWARD

TOPOGRAPHY OF ENTORHINAL PATHWAYS

Fig. 20 Following a complete entorhinal lesion (similar to the one illustrated i n fig. l ) , degenerating fibers can also be seen to emerge from the dorsal psalterium (psd) both ipsilateral and contralateral to the lesion, and cross the stratum oriens, pyramidale, and radiatum to eventually join the rather dense terminal field i n the stratum lacunosum-moleculare of regio superior.

309

310

OSWALD STEWARD

Fig. 21 Organization of entorhinal afferents along the dendrites of granule cells of the area dentata ( A D ) , pyramidal cells of regio superior (RS) and regio inferior ( R I ) .

rial ('34). The present report may therefore represent the first demonstration of Lorente de No's alvear pathway in experimental material. The rostral continuation of the projection to the portion of regio superior furthest from the CA1-CAB transition seems also to follow a rather unique route. In the rostral hippocampal formation (as viewed in coronal section), lesions or injections which involve either pars lateralis or pars intermedialis reveal a component of degenerating fibers which emerges from the dorsal psalterium near the point where CA1 makes a flexure (fig. 19A). These fibers then cross the stratum pyramidal, and radiatum, and join the dense terminal field in the stratum lacunosum-moleculare of CA1 (fig. 20B). A similar fiber com-

ponent can also be seen in the CA1 region contralateral to the lesion. DISCUSSION

A schematic summary of the results of the present study is illustrated in figure 21. Two distinctly different organizational features characterize the projections from medial and lateral entorhinal regions. First, the pathways to the ipsilateral fascia dentata and regio inferior from the various medio-lateral divisions of the entorhinal area distribute along the dendrites of a single cell (dentate granule cells or pyramidal cells of regio inferior respectively). Medially originating afferents terminate proximal to the cell somata, while laterally originating afferents terminate on more distal dendritic segments. The projections

TOPOGRAPHY OF ENTORHINAL PATHWAYS

31 1

to regio superior are, however, organized longitudinally across a field of dendrites with fibers from the lateral-most portion of the entorhinal area projecting to the stratum lacunosum-moleculare of CA1 closest to the subiculum and stratum moleculare of the subiculum proper, and fibers from the medial entorhinal area terminating in the portion of CA1 furthest from the subiculum. This longitudinal organization of entorhinal projections is maintained throughout the rostro-caudal axis of the hippocampus, and also in the terminal field of the crossed temporo-ammonic tract, although in caudal regions the crossed teniporo-ammonic projections from lateral entorhinal regions to the CAl-subicular zone tend to decrease in density.

report that in the post-lesion reorganized state, medial and lateral entorhinal pathways distribute along the dendrites of the granule cells of the contralateral dentate gyrus in a manner which is highly reminiscent of the normal laminated pattern of projection. This specificity i n the post lesion state could come about either because the sprouting fibers simply proliferated within their normal terminal zones, or if a significant new population of axons grew into the denervated stratum moleculare, they could simply follow the few normal "pioneer" fibers already in place. A clear distinction between these possibilities must await a detailed description of the sequence of proliferative alterations i n the crossed temporo-dentate projections.

Projections to the dentate gyrus and regio inferior These observations provide a conception of entorhinal projections which for the dentate gyrus and regio inferior is not substantially different from that of HjorthSimonsen and Jeune ('72), and HjorthSimonsen ('72). These authors suggest that the medio-lateral divisions of the entorhinal area could either give rise to two discrete pathways (a medial and a lateral), two pathways with a transition zone interposed (medial, transition, and lateral) or a graded transition in the pattern of termination of lateral and medial projections. The present results suggest a graded transition since at least four patterns of lamination were observed which seemed to overlap considerably with projections from adjacent medio-lateral divisions of the entorhinal area. However, the results are not inconsistent with the interpretation that there are multiple discrete subareas in the entorhinal region, and that within and between each subdivision, the patterns of projection are graded in a mediolateral dimension. A comparable laminated distribution of medial and lateral entorhinal afferents also exists i n the normal crossed projections to the contralateral dentate gyrus. This specificity i n the normal state provides a ready explanation for the specificity in the pattern of termination following lesion-induced proliferation of the crossed projections. Steward et al. ('74b), and Zimmer and Hjorth-Simonsen ('75)

Projections to regio superior The present observations also add a new dimension to the understanding of the topographic organization of entorhinal projections to regio superior. This pathway has been the subject of some disagreement in the literature, since some authors report heavy degeneration within the stratum lacunosum-moleculare of CA1 following entorhinal lesions (Blackstad, '56, '58; Raisman et al., '65; and Steward et al.,'73, '74), while other investigators report a negligible projection (notably Naf st ad, '67). The reasons for these discrepancies may now be understood in terms of the topographic organization of the projection. In discussiong the projection, one must specify the medio-lateral segment of the entorhinal area, and the site in regio superior along the subicular-CA2 axis. With complete medial and lateral entorhinal lesions, the complete pattern of termination may be defined as extending from the CA1CA2 transition zone to the stratum moleculare of the subiculum proper (Steward et al., '73, '74; and figs. 17-19). With more selective lesions, such as those utilized by Nafstad ('67), the terminal field would be restricted to only a portion of this total projection area. The very medial lesions utilized by Nafstad would be likely to result in a pattern of terminal degeneration comparable to the pattern illustrated autoradiographically in figure 17A, but such a medially placed lesion would result in virtually no terminal degeneration in that portion of the CA1 stratum lacuno-

312

OSWALD STEWARD

sum-moleculare analyzed electron microscopically by Nafstad (see his fig. 1 ), a segment midway along the CA2-subicular axis of regio superior. Indeed, Nafstad reports that only 1% of the terminals within this region degenerate following medial entorhinal lesions, a finding that is not inconsistent with the observations of the present paper. The present experiments also have resulted in the first experimental demonstration in the rat of a fiber system having the characteristics of Lorente de No's alvear path ('34). Lesions or injections which include the lateral-most subdivision of the entorhinal area, reveal a component of fibers in the alveus which give rise to a fiber component which traverses the stratum pyramidale of the subiculum, en route to the deeper stratum moleculare. In both the stratum pyramidale and stratum moleculare of the subiculum, silver staining reveals apparent terminal arborization, but the terminal degeneration is much more dense i n the stratum moleculare. While the present experiments reveal the presence of a n entorhinal projection system having the characteristics of the alvear path, i t is clear that this system is only the most laterally originating component of a more extensive projection system to both the subiculum and the CA1 region, and does not represent a separate pathway. It has been suggested that the periamygdaloid region, possibly including the prepyriform cortex and the amygdala proper give rise to some afferents to the CA1subicular transition region (Cragg, '61; Powell et al., '65; and Krettek and Price, '74). In these cases, the projections to the CAl-subicular zones were observed in the absence of any projection to the stratum moleculare of the fascia dentata, suggesting that at least some of the projections to CA1 and the subiculum originate from areas beyond the most lateral entorhinal zones which give rise to temporo-dentate projections. The present results are not inconsistent with this interpretation, but do again suggest that if such a projection exists, i t may be considered as a part of the larger medio-lateral organization in the pattern of afferent projection to CA1. Of particular interest in this regard is the demonstration in a subsequent com-

munication (Steward and Scoville, '76) that afferents to CA1 and subiculum arise from layer 111 pyramidal cells in the entorhinal region, while afferents to the fascia dentata arise from layer I1 stellate cells. Since the layer 111 pyramidal cells seem to extend further laterally than layer I1 stellate cells (into the area called PRh in the present report) it may be that this area is selectively involved when lesions or injections reveal a terminal field in the CA1subicular zone, and not in the fascia dentata. This question will require much more detailed analyses, however. After the present paper was submitted, a report appeared on the medio-lateral organization of entorhinal efferents in the rhesus monkey which corroborates some of the present observations in the rat (Van Hoesen and Pandya, '75b). In particular, their observations on the medio-lateral organization of entorhinal afferents to the fascia dentata are quite consistent with the present report. These authors demonstrated that pars medialis of the entorhinal area (area 28a) projects to the middle portions of the stratum moleculare of the fascia dentata, while pars lateralis (area 28b) projects to the outer most portions of the stratum moleculare. In addition, while they were not able to selectively ablate pars intermedialis (area 28i) they predicted, on the basis of lesions which also included other entorhinal regions, that this intermediate zone might project to an intermediate location in the dentate stratum moleculare. With regard to the organization of the projections to regio inferior and superior of the hippocampus proper, however, their observations are not totally in accord with those of the present study. First, they did not observe a concentric laminated pattern of termination of entorhinal afferents in regio inferior, such as that illustrated in the present material (see figs. 15, 21). Second, the topographic organization of entorhinal afferents in regio superior was also somewhat different. Both pars medialis and pars lateralis (areas 28a and 28b) were found to project to a portion of CA1 relatively near the CA1-CA2 transition. In addition, these two areas seemed to terminate on different proximo-distal portions of the CA1 pyramidal cell dendrites. Pars inter-

,

TOPOGRAPHY OF EN'I'ORHINAL PATHWAYS

medialis (area 28i) however, was found to terminate slightly further from the CAlCA2 transition. They suggest that their area Pr2 projects to the portions of CA1 nearest the subiculum, and to the subiculum proper (or prosubiculum, according to their terminology, and the terminology of Lorento de N6, '34). Their area Pr2 is probably not homologous with the area which has been called PRh in the present report, although their area Prl may be. Indeed, in the rat, it has not been possible to observe any cytoarchitectonic area which matches their descriptions of area Pr2 in the rhesus monkey. The final discrepancy between the reports of Van Hoesen and Pandya ('75) and the present study concerns the alvear path. In their degeneration studies, they report that there is no evidence for an alvear pathway, and suggest that what Cajal ('11) and Lorente de NO ('34) might have observed was the pathway from the presubiculum to the thalamus, which does join the alveus. Both the degeneration studies, and the autoradiographic preparations of the present report, however, suggest the existence of at least a minor fiber projection which matches the classical descriptions of the alvear pathway, and which seems to originate from pars intermedialis and pars lateralis of the entorhinal area. I n particular, the lateral entorhinal projection system which leaves the alveus to cross the stratum pyramidale and stratum moleculare of the subiculum seems to correspond almost exactly to the descriptions of the alvear pathway provided by Lorente de N6 ('34). The differences between the present report, and that of Van Hoesen and Pandya ('75b) may represent species differences between the rat and monkey, or differences in experimental techniques. Some of the differences may, however, be accounted for by the fact that the cells of origin of temporodentate afferents are a different population of cells than those which give rise to the projections to regio superior. This topic will receive detailed consideration in a subsequent report (Steward and Scoville, '76). ACKNOWLEDGMENTS

Supported in part by a n Alfred Sloan Foundation Grant (#72-11-4), and in part by USPHS Research Grant #R01 NS12333-

313

01 to the author. I thank S . 0. E. Ebbesson and L. Heimer for their helpful discussions. LITERATURE CITED

Blackstad, T. W. 1956 Commissural connections of the hippocampal region in the rat, with special reference to their mode of termination. J. Comp. Neur., 105: 417-537. - 1958 O n the termination of some afferents to the hippocampus and fascia dentata. An experimental study in the rat. Acta Anat., 35: 202-214. Cowan, W. M., D. I. Gottlieb, A. E. Hendrickson, J. L. Price and T. A. Woolsey 1972 The autoradiographic demonstration of axonal connections i n the central nervous system. Brain Res., 37: 21-51. Cragg, B. G. 1961 Olfactory and other afferent connections of the hippocampus in the rabbit, rat, and cat. Exp. Neur., 3: 588-600. Ebbesson, S. 0. E. 1970 The selective silverimpregnation of degenerating axons and their synaptic endings in non-mammalian species. In: Contemporary Research Methods in Neuroanatomy. W. J. H. Nauta and S. 0. E. Ebbesson, eds. Springer-Verlag, Heidelberg, pp. 132161. Fink, R. P., and L. Heimer 1967 Two methods for selective silver impregnation of degenerating axons and their synaptic endings in the central nervous system. Brain Res., 4: 369-374. Goldowitz, D., W. F. White, 0. Steward, G. Lynch and C. Cotman 1975 Anatomical evidence for a projection from the entorhinal cortex to the contralateral dentate gyrus of the rat. Exp. Neur., 47: 433-441. Heimer, L. 1975 A new look at the basal forebrain. Seventh Annual Pickney J. Harman Memorial Lecture. Cajal Club, Los Angeles, California. Hjorth-Simonsen, A. 1972 Projection of the lateral part of the entorhinal area to the hippocampus and fascia dentata. J. Comp. Neur., 146: 219-232. 1973 Some intrinsic connections of the hippocampus i n the rat: An experimental analysis. J. Comp. Neur., 147: 145-162. Hjorth-Simonsen, A., and B. Jeune 1972 Origin and termination of the hippocampal perforant path i n the rat studied by silver impregnation. J. Comp. Neur., 144: 215-232. Krieg, W. J. S. 1946 Connections of the cerebral cortex. I. The albino rat. A. Topography of the cortical areas. J. Comp. Neur., 84: 221276. Lorento de NO, R. 1934 Studies on the structure of the cerebral cortex. 11. Continuation of the study of the Ammonic System. J. Psychol. Neurol., 46: 113-177. Nafstad, P. H. J. 1967 A n electron microscope study on the termination of the perforant path fibers in the hippocampus and the fascia dentata. Z. Zellforsch., 76: 532-542. Powell, T. P. S., W. M. Cowan and G. Raisman 1965 The central olfactory connections. J. Anat. (London), 99: 791-813.

-

314

OSWALD STEWARD

Price, J. L., and T. P. S. Powell 1971 Certain observations on the olfactory pathway. J. Anat., 110: 105-126. Raisman, G., W. M. Cowan and T. P. S. Powell 1965 The extrinsic afferent commissural and association fibers of the hippocampus. Brain, 88: 963-996. Ramdn y Cajal, S. 1911 Histologie du Systeme Nerveux de l’homme et des vertCbr6s. T. 2. Instituto Ramon y Cajal, Madrid, 1955, 995 pp. Steward, O., C. Cotman, and G. Lynch 1973 Re-establishment of electrophysiologically functional entorhinal cortical input to the dentate gyrus deafferented by ipsilateral entorhinal lesions: innervation by the contralateral entorhinal cortex. Exp. Brain Res., 18: 396-414. - 1974a Growth of a new fiber projection in the brain of adult rats: re-innervation of the dentate gyrus by the contralateral cortex following ipsilateral entorhinal lesions. Exp. Brain Res., 20: 45-66. 1974b Selectivity in the pattern of new synapse formation with denervated dentate

granule cells. Fourth Annual Meeting of the Society for Neuroscience. St. Louis, Missouri, (abstracts ). Steward, O., and S. A. Scoville 1976 Cells of origin of entorhinal cortical afferents to the hippocampus and fascia dentata of the rat. J. Comp. Neur., in press. Van Ferreira, A. 1951 The cortical areas of the albino rat studied by silver impregnation. J. Comp. Neur., 95: 177-243. Van Hoesen, G . W., and D. N. Pandya 1975a Some connections of the entorhinal (area 28) and perirhinal (area 35) cortices of the rhesus monkey. I. Temporal lobe afferents. Brain Res., 95: 1-24. 1975b Some connections of the entorhinal (area 28) and perirhinal (area 35) cortices of the rhesus monkey. 111. Efferent connections. Brain Res., 95: 35-59. Zimmer, J., and H. Hjorth-Simonsen 1975 Crossed pathways from the entorhinal area to the fascia dentata: 11. Provokable i n rats. J. Comp. Neur., 161: 71-102.

Topographic organization of the projections from the entorhinal area to the hippocampal formation of the rat.

The present study re-examines, with autoradiographic methods, the pattern of termination of fibers originating from various medio-lateral divisions of...
9MB Sizes 0 Downloads 0 Views