Difference between revisions of "Rat Hippocampus Atlas"

From projects.navigator3
Jump to: navigation, search
Line 704: Line 704:
 
'''Structures associated with this boundary:&nbsp;'''<br/>[[#Postrhinal_cortex|Postrhinal cortex (POR)]]<br/>Visual/Temporal Cortex
 
'''Structures associated with this boundary:&nbsp;'''<br/>[[#Postrhinal_cortex|Postrhinal cortex (POR)]]<br/>Visual/Temporal Cortex
  
On the basis of cytoarchitectonic criteria, the dorsal border of the postrhinal cortex (POR) is difficult to establish throughout its extent. POR lacks a clear laminar structure which is more conspicous in the adjacent cortex. Layer II of temporal and primary visual cortex is more densely packed, and cells in layers II and III tend to be slightly smaller than in POR. Layer V in POR also is less dense compared to adjacent cortex and layer IV is less conspicuous in POR. Ectopic layer II cells, a distinct feature of POR are absent in the dorsally adjacent cortices. Using immunostaining for calbindin does not show much of a border although at the dorsal border of POR, there is a slight increase in staining intensity in deeper layers, resulting in a more bilaminar staining pattern. Sections stained for parvalbumin though provide a pretty strict border between since the dorsally adjacent cortex does stain with this marker whereas all layers of POR are almost completely negative except for the presence of an occasional positive neuron.&nbsp;
+
On the basis of cytoarchitectonic criteria, the dorsal border of the postrhinal cortex (POR) is difficult to establish throughout its extent. POR lacks a clear laminar structure which is more conspicous in the adjacent cortex. Layer II of temporal and primary visual cortex is more densely packed, and cells in layers II and III tend to be slightly smaller than in POR. Layer V in POR also is less dense compared to adjacent cortex and layer IV is less conspicuous in POR. Ectopic layer II cells, a distinct feature of POR are absent in the dorsally adjacent cortices. Using immunostaining for calbindin does not show much of a border although at the dorsal border of POR, there is a slight increase in staining intensity in deeper layers, resulting in a more bilaminar staining pattern. Sections stained for parvalbumin though provide a pretty strict border between since the dorsally adjacent cortex does stain with this marker whereas all layers of POR are almost completely negative except for the presence of an occasional positive neuron.&nbsp;<br/><br/>
  
 
=== Presubiculum (PrS) / Subiculum (SUB) ===
 
=== Presubiculum (PrS) / Subiculum (SUB) ===

Revision as of 13:10, 16 February 2015

Welcome to the Rat Hippocampus Atlas - Version 1.2

  • The rat hippocampus atlas is an interactive resource providing:
  • Detailed descriptions of 18 hippocampal structures
  • Criteria for delineating 63 boundaries between these structures
  • Histological image repository with boundaries delinated according to the described cyto- and chemoarchitectonic criteria
  • Bidirectional links between the structure/boundary descriptions (Index of structures) and the corresponding section images (Image viewer)


The system is intended for researchers working in the field, as well as students interested in this brain region. The atlas is accessed through the structure index or image viewer. See tutorial for details about functionality. These web pages contain Flash applications (see requirements)

 

Re-use of data from this repository is allowed provided that reference is given to the following publication:

Kjonigsen LJ, Leergaard TB, Witter MP, Bjaalie JG. 2011. Digital atlas of anatomical subdivisions and boundaries of the rat hippocampal region. Frontiers in Neuroinformatics, 2011; 5:2. doi: 10.3389/fninf.2011.00002


Contents

News

2011-03 Speed optimization. Release of version 1.2

2011-01 Modified version, released for review. Release of version 1.1

2010-11 Deployed for review in conjunction with submitted original manuscript

2010-06 Poster, 7th Forum for European Neuroscience 2010, Amsterdam, The Netherlands

2009-11 Poster / demonstration, Society of Neuroscience conference in Chicago, USA

2009-09 Poster / demonstration, 2nd INCF Congress of Neuroinformatics, Plzen, Czech Republic

 


External links

The Hippocampal Connectome Project

(van Strien NM, Cappaert NL, Witter MP (2009). The anatomy of memory: an interactive overview of the parahippocampal-hippocampal network. Nat Rev Neurosci.10(4): 272-82)


3D

 

Image viewer

The image repository consists of ~100 coronal images stained for NeuN, calbindin, and parvalbumin. The triple viewer allows interactive zooming and panning of each viewer panel, with a possibility to synchronize the position and zoom in the three panels. Annotations can be toggled on / of, and pre-set high-magnification positions for annotated regions are available. For further details, see the tutorial.


Image repository :

Name

Species

CuttingDirection

Comments

Hippocampus atlas application (coronal)  

RAT 

CORONAL 

NeuN, parvalbumin and calbindin 

 

Tutorial

The atlas system has two main elements:

  • An illustrated text describing 18 hippocampal structures and 63 boundaries between these structures. The structure descriptions can be accessed from an alphabetical or hierarchical index.
  • An image repository of ~100 coronal images showing the cyto- and chemoarchitecture of the hippocampus region. The images are organized in triplets of neighbouring sections stained for calbinding, NeuN, and parvalbumin. The image triples are sorted from anterior to posterior by distance from bregma.

The structure and boundary definitions and the image viewer open in separate browser windows, and bidirectional links are provided between corresponding structure descriptions and image levels. Tools for querying are not yet implemented. For optimal use it is recommended to view the browser windows in full screen, preferably using a dual screen setup.

To use the index of structures, scroll through the alphabetical index to a structure name of interest. Links are provided to structure and boundary descriptions as well as to corresponding levels in the image repository. The structure descriptions give a review of key anatomical features, and provides links to descriptions of boundaries with surrounding structures and corresponding levels in the image repository. The boundary descriptions review the characteristic histological criteria used to identify the boundaries in histological sections, together with links to corresponding levels in the image repository.


(MUST BE REVISED) To use the image viewer first drag the slider under the row of thumbnail images horizontally to select an anterioposterior level of interest (Bregma coordinates are indicated). Click on the thumbnail images to open them in the triple viewer. The image panels show images of sections stained for calbindin (left), NeuN (middle), and parvalbumin (right). Not all image triplets are complete. The viewer has the following options for navigating the images:

·         Zoom, click the + button in the toolbar or press the shift key to zoom in; click the - button or press the ctrl key to zoom out

·         Pan, click the arrows in the tool bar, or left-click and drag the images freely in the viewer

·         Synchronize, to synchronize the three viewer panels to the same position and zoom scale, click the = button in the toolbar

·         Pre-set, to navigate to a predefined view of a given region, select a structure in the Annotation toolbar and click GO

·         Reset, to reset viewer to default, click the reset button in the toolbar

·         Annotation, click the annotation button in the toolbar to toggle annotations on and off. Graphical overlay is only available for the NeuN images in the middle viewer panel. The employed color coding is in correspondence with the interactive connectivity diagram provided by van Strien et al., 2009; Nat Rev Neurosci.10:272-282

·         Scale bar, the ZoomScale bar shows a scale bare for the image panels. The bar is transparent and can be moved to make crude measurements in the images

To look up structure descriptions and boundary definitions, click on the color-coded structure name abbreviations in the top panel.


Examples of use:

The atlas system is designed to provide both text description and annotated histological details from the rat hippocampus. The atlas has two main elements, which serve as entry points for use:

·         Find a structure of interest using the alphabetical or hierarchical structure index

·         Find a structure or position of interest using the image repository and the provided stereotaxic positions (bregma levels).

A student of the hippocampal region may for example use the structure index as a starting point to read about the structure, and then explore the histoarchitecture in the image repository. A researcher seeking to define a boundary in experimental histological material, may for example look up a corresponding level in the image repository, and then read about boundary definitions through the index of structures.


 

About the project

The rat hippocampus atlas is an interactive resource providing a systematic overview of cyto- and chemoarchectonical feastures of the hippocampus proper, fasciola, and associated parahippocampal cortices. This atlas system has been developed to serve the need to integrate detailed descriptions of structures and criteria defining boundaries and atlas images in which the underlying histological features can be explored.


Features

  • Alphabetical and hierarchical overview of 18 hippocampal structures
  • Detailed, illustrated descriptions of 63 boundaries
  • Interactive image repository with ~100 coronal histological images stained for NeuN, calbindin, and parvalbumin
  • Triple image viewer in wich differently stained neighbouring sections can be interactively compared
  • Graphical overlay of substructures based on described boundary criteria
  • Bidirectional links between structure descriptions and image repository


Methods

The atlas is based on histological material from an adult Long Evans rat, stained for NeuN, calbindin, and parvalbumin (for details, see Kjonigsen et al., Frontiers in Neuroinformatics, 2011; 5:2. doi: 10.3389/fninf.2011.00002).


Contact

Digital brain atlasing:

j.g.bjaalie@medisin.uio.no

Technical issues and functionality:

i.a.moene@medisin.uio.no

Neuroanatomy of the hippocampus:

menno.witter@ntnu.no


Contributing laboratories:

Kavli Institute for Systems Neuroscience & Centre for the Biology of Memory

Medical-Technical Research Centre 
Norwegian University of Science and Technology 
N - 7489 Trondheim 
Norway 
http://www.ntnu.no/cbm/

Neural Systems and Graphics Computing Laboratory 
Institute of Basic Medical Sciences 
Department of Anatomy 
University of Oslo 
P.O. Box 1105 Blindern 
N - 0317 Oslo 
Norway 
http://www.nesys.uio.no


Credits

NeSys and CMBN atlas development team:

Lisa J. Kjonigsen, M.Sc.
Jan Olav Kjode, M.Sc.
Ivar Andre Moene, M.Sc.
Trygve B. Leergaard, M.D., Ph.D.
Jan G. Bjaalie, M.D., Ph.D.

KI-CBM and Witter-group:
Menno P. Witter, Ph.D.


Documentation

The Rat Hippocampus Atlas System: 
Kjonigsen LJ, Leergaard TB, Witter MP, Bjaalie JG. 2011. Digital atlas of anatomical subdivisions and boundaries of the rat hippocampal region. Frontiers in Neuroinformatics, 2011; 5:2. doi: 10.3389/fninf.2011.00002 

Hippocampal boundary definitions, ontology and nomenclature:
Boccara CN, Kjonigsen LJ, Hammer I, Bjaalie JG, Leergaard TB, Witter M.P. Manuscript in preparation 

Database architecture:
Moene IA, Subramaniam S, Darine D, Leergaard TB, Bjaalie JG. 2007. Towards a workbench for rodent brain image data: systems architecture and design. Neuroinformatics; 5: 35-58


References for the index of structures

Amaral DG, Scharfman HE, Lavenex P. 2007. The dentate gyrus: fundamental neuroanatomical organization (dentate gyrus for dummies) Prg Br Res163:3-22.

Blackstad TW. 1956. Commissural connections of the hippocampal region in the rat, with special reference to their mode of termination. J Comp Neurol 105:417-519.

Brodmann K. 1909. Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbauers. Leipzig.

Burwell RD. 2001. Borders and cytoarchitecture of the perirhinal and postrhinal cortices in the rat. J Comp Neurol 437(1):17-41.

Caballero-Bleda M, Witter MP. 1993. Regional and laminar organization of projections from the presubiculum and parasubiculum to the entorhinal cortex: an anterograde tracing study in the rat. J Comp Neurol 328(1):115-129.

Haug FM. 1974. Light microscopical mapping of the hippocampal region, the pyriform cortex and the corticomedial amygdaloid nuclei of the rat with Timm's sulphide silver method. I. Area dentata, hippocampus and subiculum. Z Anat Entwicklungsgesch 145(1):1-27.

Haug FM. 1976. Sulphide silver pattern and cytoarchitectonics of parahippocampal areas in the rat. Special reference to the subdivision of area entorhinalis (area 28) and its demarcation from the pyriform cortex. Adv Anat Embryol Cell Biol 52(4):3-73.

Insausti R, Herrero MT, Witter MP. 1997. Entorhinal cortex of the rat: cytoarchitectonic subdivisions and the origin and distribution of cortical efferents. Hippocampus 7(2):146-183. 

Jaffe DB, Gutierrez R. 2007. Mossy fiber synaptic transmission: communication from the dentate gyrus to area CA3. Prg Br Res 163:109-132. 

Krettek JE, Price JL. 1977. Projections from the amygdaloid complex and adjacent olfactory structures to the entorhinal cortex and to the subiculum in the rat and cat. J Comp Neurol 172(4):723-752. 

Paxinos G, Watson C. 1998. The Rat Brain In Stereotaxic Coordinates: Academic Press. 

Ruth RE, Collier TJ, Routtenberg A. 1982. Topography between the entorhinal cortex and the dentate septotemporal axis in rats: I. Medial and intermediate entorhinal projecting cells. J Comp Neurol 209(1):69-78. 

Ruth RE, Collier TJ, Routtenberg A. 1988. Topographical relationship between the entorhinal cortex and the septotemporal axis of the dentate gyrus in rats: II. Cells projecting from lateral entorhinal subdivisions. J Comp Neurol 270(4):506-516. 

Stephan H. 1975. Allocortex. Handbuch der Mikroskopischen Anatomie des Menschen. Berlin: Springer-Verlag.

Steward O. 1976. Topographic organization of the projections from the entorhinal area to the hippocampal formation of the rat. J Comp Neurol 167(3):285-314.

van Groen T, Wyss JM. 1990. The postsubicular cortex in the rat: characterization of the fourth region of the subicular cortex and its connections. Brain Res 529(1-2):165-177.

van Strien NM, Cappaert NL, Witter MP. 2009. The anatomy of memory: an interactive overview of the parahippocampal-hippocampal network. Nat Rev Neurosci. 10 (4):272-282.

Witter MP, Amaral DG, Paxinos G. 2004. Hippocampal formation. The Rat Nervous System. San Diego, CA: Elsevier Academic Press. p 635-704. 

Wyss JM. 1981. An autoradiographic study of the efferent connections of the entorhinal cortex in the rat. J Comp Neurol 199(4):495-512.


Release notes

Version 2.0

- Text describing structures and borders updated

- Transfer to wiki

- Image viewer updated

- Sagital and horizontal sections added to image viewer 

Version 1.2
- optimized the data upload to the viewer.

Version 1.1
- better support for Mozilla Firefox 
- some changes in the text 
- links are added in the structure descriptions


 

Structure descriptions

Amygdalo-entorhinal transition area (AE)

Boundaries associated with this structure: 

Amygdalo-entorhinal transition area (AE) / Dorsal Intermediate Entorhinal Area (DIE)
Amygdalo-entorhinal transition area (AE) / Periamygdaloid cortex
Amygdalo-entorhinal transition area (AE) / Ventral Intermediate Entorhinal Cortex (VIE)


This field is located between the ventral-intermediate entorhinal area (VIE) and the amygdalohippocampal transition area. The former is caudal and the latter is anterior to AE. Laterally, AE borders with the dorsal-intermediate entorhinal area (DIE), while medially it abuts on the periamygdaloid cortex. 

The lamination of this field is less developed than in the remainder of entorhinal fields. Layer I is very broad and has a scalloped border with layer II. Layers II and III merge, and are made up of small-sized cells. There is a distinct tendency for cells in layers II and III to form clusters, leaving broad gaps in between. Layer V is characterized by the presence of a population of slightly larger neurons, more homogeneous in appearance than layer III. Layer VI is mainly made up of a variety of loosely arranged neurons such that no clear border with the white matter can be recognized. Field AE is largely comparable to the field TR as described by Haug (1976), which is similar to the area semi-annularis according to Stephan (1975). 

 
 

Caudal entorhinal area (CE)

Boundaries associated with this structure: 

Dorsal intermediate entorhinal area (DIE)
Dorsolateral entorhinal area (DLE)
Medial entorhinal area (ME)
Ventral intermediate entorhinal area (VIE)


The caudal entorhinal area makes up the ventral part of the caudal pole of the rat cerebral hemisphere. At the very caudal extreme, it spans the mediolateral extent of the entorhinal cortex, from the rhinal fissure laterally to the border with the parasubiculum medially. CE has a strikingly columnar appearance. Layer II forms a continuous band of medium- or big-sized, darkly stained neurons, sharply delimited from layer I. At its medial extreme, layer II suddenly increases its width, forming a club-like thickening. Layer III contains medium-sized pyramidal cells that are homogeneously distributed. Layers III and V are separated by a clear cell free zone or lamina dissecans. Layer V is populated by medium-sized neurons that are loosely arranged at the border with the lamina dissecans, whereas in the deep part of the layer, these cells tend to be somewhat more tightly arranged. Layer VI is broad and not clearly demarcated from the white matter. In layers V and VI, neurons are oriented in a columnar fashion. 

The architectonic features of the caudal entorhinal field, in particular those of the superficial layers II and III, and the presence of a clearly demarcated lamina dissecans are largely coincident with the descriptions of the typical medial entorhinal cortex by previous authors (Blackstad, 1956; Haug, 1976; Steward, 1976; Wyss, 1981; Ruth et al., 1982). However, the extent of the medial entorhinal cortex as defined by these authors is much larger than in our observations, and includes the medial entorhinal field (ME). 

The lateral and caudal borders between CE and the perirhinal and postrhinal cortices is indicated by: 

1. the lamina dissecans disappearing 
2. the part of CE at the border stains densely for parvalbumin whereas the adjacent perirhinal and postrhinal cortex are almost devoid of positive staining 
3. the latter areas stain for calbindin which is much less conspicuously present in the adjacent parts of EC 
4. layer II of EC is characterized by a population of large to medium sized neurons that stain very densely for neuronal markers such as Nissl or NeuN. The adjacent parts of perirhinal and postrhinal cortices are characterized by blended layers II and III consisting of small, lightly stained neurons 
5. at medial levels, CE can occasionally border a laterally extending part of PaS. This area is described below. This feature is in the rat brain quite variable and therefore animal specific; such individual variation is not commonly seen in rats

The anterior border of CE with ME is indicated by: 

1. the rather homogeneous layer II becomes less obvious and generally breaks into two or three clusters of cells 
2. layer II is separated from layer III by a narrow, irregular cell sparse zone 
3. layer III looses its very regular density and can be subdivided in an outer more densely packed zone and an inner less dense zone 
4. the superficial part of layer V shows a population of large, darkly stained pyramidal cells not present in CE 

The anterior border of CE with the ventral-intermediate entorhinal area (VIE) is indicated by: 

1. loss of the conspicuous lamina dissecans 
2. loss of the very regular appearance of layer III and the columnar appearance of layers V and VI 
3. a sudden increase in large darkly stained pyramidal cells in layer V 
4. sudden thinning of layer II 

The medial border with the parasubiculum features: 

1. a striking merge between layers II and III in the parasubiculum 
2. striking decrease in staining intensity for calbindin throughout the neuropil of all layers 
3. dense staining for acetylcholinesterase in layers I-III in parasubiculum  
 

CA1

Boundaries associated with the CA1: 

CA2
Fasciola 
Subiculum (SUB)


The Cornu Ammonis or Ammon's Horn (CA) also referred to as hippocampus proper, is characterized by a thin layer of densely packed pyramidal cells, enclosed by an outer plexiform layer and an inner polymorph layer, also called stratum oriens. The outer plexiform layer, which contains the apical dendrites of the pyramidal cells, is generally subdivided into a number of sublayers. These will be detailed in the descriptions of its three main subdivisions, area CA1, CA2, and CA3 (according to some accounts a field CA4 can be differentiated as well, but because this is ill-defined with respect to the hilus of the dentate gyrus, this practice will not be followed here (Witter, 2004). The CA field is bordered on one side by the dentate gyrus (DG); this side of CA is generally referred to as the proximal part. The opposite distal side is where CA meets the subiculum (SUB). The CA, and thus its constituting subfields, is a three-dimensionally complex structure such that its orientation is often difficult to grasp when seen in different planes of sectioning.

Area CA1 is part of the hippocampus proper, whereas the subiculum is generally considered to represent a separate entity. However, both areas share the layered composition which is typical for hippocampal fields. This layered apprearance is most easily seen in Nissl, NeuN or Timm-stained sections. In CA1, and less prominently in the subiculum, a wide molecular layer is located between the pyramidal layer and the hippocampal fissure. In the subiculum, this molecular layer can be subdivided into a deeper portion that is continuous with the stratum radiatum of CA1, and a superficial portion that is continuous with stratum lacunosum-moleculare. Note that this subdivision is difficult to detect by the markers used in this atlas. The superficial molecular layer of the subiculum and its continuation in stratum lacunosum moleculare of CA1 contain the perforant pathway fibers from the EC. Moreover, afferents from other structures such as the nucleus reuniens of the midline thalamus take a similar course and distribution. 

CA1 can be easily recognised in Nissl and NeuN-stained sections due to its layer of neatly aligned pyramidal cells. This feature distinguishes CA1 from the subiculum, which dorsally is located between area CA1 and the retrospenial cortex, and ventrally between area CA1 and pre- and parasubiculum. The CA1/subiculum border is clearly marked by an abrupt widening of the pyramidal cell layer. Moreover, in material stained for parvalbumin or AChE, the pyramidal layer of CA1 is darkly stained, whereas the pyramidal cell layer of the subiculum is more diffusely stained, thus indicating a marked border between the two fields. In the TIMM staining the border is also visible, but then with an unstained CA1pyramidal layer, and a darkly stained cell layer in the proximal part of the subiculum. Although the border may appear easy to establish, it is in practice not possible to determine whether a particular dendrite at the border belongs to a neurone in CA1 or in the subiculum. This is caused by the oblique orientation of the CA1 / subiculum border relative to the transverse axis. Therefore, cells (in the stratum oriens, stratum radiatum, or lacunosum moleculare) close to this border do not necessary extent their dendrites perpendicularly to the orientation of stratum pyramidale.  
 

CA2

Boundaries associated with this the CA2: 

CA1
CA3


The Cornu Ammonis or Ammon's Horn (CA) also referred to as hippocampus proper, is characterized by a thin layer of densely packed pyramidal cells, enclosed by an outer plexiform layer and an inner polymorph layer, also called stratum oriens. The outer plexiform layer, which contains the apical dendrites of the pyramidal cells, is generally subdivided into a number of sublayers. These will be detailed in the descriptions of its three main subdivisions, area CA1, CA2, and CA3 (according to some accounts a field CA4 can be differentiated as well, but because this is ill-defined with respect to the hilus of the dentate gyrus, this practice will not be followed here (Witter, 2004). The CA field is bordered on one side by the dentate gyrus (DG); this side of CA is generally referred to as the proximal part. The opposite distal side is where CA meets the subiculum (SUB). The CA, and thus its constituting subfields, is a three-dimensionally complex structure such that its orientation is often difficult to grasp when seen in different planes of sectioning.

The CA2 and CA3 parts of the Ammon's horn are characterized by the presence of large pyramidal cells, forming a few closely packed layers on top of one another. Both have a layered appearance that is shared with CA1 and the subiculum. Below the pyramidal cell-layer, a relatively cell-free layer is seen, the so-called stratum oriens. Below the stratum oriens the fibre tract alveus can be found. Directly superficial to the pyramidal cell layer of CA1, the stratum radiatum is situated (darkly stained in TIMMs stain). Between the stratum radiatum and the hippocampal fissure, the stratum lacunosum-moleculare is found (not stained with the TIMM-stain, but appearing darker than the stratum radiatum in material stained for AChE). These superficial layers, the stratum radiatum and lacunosum-moleculare contain the apical dendritic trees of the pyramidal cells.

 
 

CA3

Boundaries associated with the CA3: 

CA2
Dentate gyrus (DG)

The Cornu Ammonis or Ammon's Horn (CA) also referred to as hippocampus proper, is characterized by a thin layer of densely packed pyramidal cells, enclosed by an outer plexiform layer and an inner polymorph layer, also called stratum oriens. The outer plexiform layer, which contains the apical dendrites of the pyramidal cells, is generally subdivided into a number of sublayers. These will be detailed in the descriptions of its three main subdivisions, area CA1, CA2, and CA3 (according to some accounts a field CA4 can be differentiated as well, but because this is ill-defined with respect to the hilus of the dentate gyrus, this practice will not be followed here (Witter, 2004). The CA field is bordered on one side by the dentate gyrus (DG); this side of CA is generally referred to as the proximal part. The opposite distal side is where CA meets the subiculum (SUB). The CA, and thus its constituting subfields, is a three-dimensionally complex structure such that its orientation is often difficult to grasp when seen in different planes of sectioning.

These two parts of the Ammon's horn are characterized by the presence of large pyramidal cells, forming a few closely packed layers on top of one another. Both have a layered appearance that is shared with CA1 and the subiculum. Below the pyramidal cell-layer, a relatively cell-free layer is seen, the so-called stratum oriens. Below the stratum oriens the fibre tract alveus can be found. Directly superficial to the pyramidal cell layer of CA1, the stratum radiatum is situated (darkly stained in TIMMs stain). Between the stratum radiatum and the hippocampal fissure, the stratum lacunosum-moleculare is found (not stained with the TIMM-stain, but appearing darker than the stratum radiatum in material stained for AChE). These superficial layers, the stratum radiatum and lacunosum-moleculare contain the apical dendritic trees of the pyramidal cells.

 

Dentate gyrus (DG)

Boundaries associated with this structure: 

CA3


The dentate gyrus is characterized by the presence of a curved cell layer, densely packed with granule cells, i.e. cells that lacks basal dendrites but have an extensive apical dendritic tuft, extending into he molecular layer. The curved cell layer presents itself as a V-shaped, or C-shaped structure, depending on the dorsoventral level of the hippocampal formation (HF). Here, the portion of the granule cell layer adjacent to CA1, is referred to as the enclosed blade (which is synonymous with suprapyramidal, dorsal, or inner blade/limb), while the opposite portion of the granule cell layer is called the exposed blade (which is synonymous with infrapyramidal, ventral, outer, or free blade/limb).

The granule cell layer is covered by a molecular layer and it encloses a polymorph layer, referred to as the hilus. The overall cytoarcitecture of DG is easily appreciated in NeuN or Nissl stains. Other staining methods do not add information to specifically delineate DG, although with a stain for calbindin, the hilus is clearly set apart from the adjacent proximal part of CA3 as a positive area, similar to the stained mossy fibers that emanate from DG reaching cells in the hilus as well as providing the main connection to CA3. This staining pattern is mimicked by staining with an antibody against dynorphin. 
 

Dorsal intermediate entorhinal area (DIE)

Boundaries associated with this structure:

Caudal entorhinal field (CE)

Dorsal-lateral entorhinal area (DLE)

Ventral-intermediate entorhinal area (VIE)


The DIE field forms the ventrolateral portion of the entorhinal cortex. Layer I is rather narrow and contains very few scattered neurons. Layer II contains rather big, rounded neurons that are distinctly stained in regular Nissl stain. A very narrow, relatively acellular band separates layer II from layer III in much of the DIE. Layer III is wide, and clearly presents a narrow, more densely packed outer zone with its neurons arranged in cell clusters, and a less densely and irregularly packed inner zone (Krettek & Price, 1977).

According to several authors (cf. Caballero-Bleda and Witter, 1993) the outer portion of layer III of field DIE should be considered part of layer II. This deep portion of layer II is in such instances designated as layer IIb, whereas the more superficial portion is named IIa. Layer IV is very poorly developed or absent, so layer III abuts on layer V. Layer V is rather narrow and comprises loosely arranged medium- or big-sized pyramids that are not as darkly stained as those in the ventral-intermediate entorhinal area (VIE) and the medial entorhinal area (ME). This difference is particularly striking in layer Va of these areas. Whereas layer Va is a more or less continuous row of large, darkly stained neurons in ME and to a slightly lesser extent in VIE, such cells are only incidentally present in DIE.

Layer VI is narrow, lacks a columnar arrangement, and is more compact than layer V. This field largely coincides with the classical lateral entorhinal cortex of Blackstad (1956; see also Steward, 1976; Wyss, 1981; Ruth, 1982, 1988), area 28L of Haug (1976), and ventromedial portions of DLEA as described by Krettek and Price (1977). 
 

Dorsolateral entorhinal area (DLE)

Boundaries associated with this structure: 

Caudal entorhinal field (CE)
Dorsal-intermediate entorhinal area (DIE)
Perirhinal Cortex (PER)


The dorsal-lateral entorhinal area (DLE) forms a strip of cortex closely related to the rhinal fissure, and therefore it is the only entorhinal field entirely located on the lateral aspect of the rat’s cerebral hemisphere. At caudal levels it occupies both banks of the rhinal fissure, while anteriorly it lies medial or ventral to the rhinal fissure. It is positioned between the perirhinal cortex dorsally and the dorsal-intermediate entorhinal area (DIE) ventrally.

Similar to DIE, DLE extends along most of the rostrocaudal extent of the entorhinal cortex, as is seen in coronal sections. Layer I is thin, and differs characteristically from what is seen in all other subdivisions of the entorhinal cortex in that it is populated by neurons that look like displaced layer II cells. Layer II is rather thin and densely packed with darkly stained elongated, big neurons. The long axis of many of these neurons runs parallel to the outer surface of the ventral bank of the rhinal fissure. Layer III is also thin, and the cells are organized in horizontal rows, parallel to the surface curvature of the rhinal fissure. The outer portion of this layer has a higher cell density than the inner one. A cell sparse layer IV separates layers III and V. 

Layer V has a few very big and darkly stained neurons, while the remaining neurons are medium-sized. Layer VI is more compact than layer V, and its cells larger than the layer VI cells of the immediately adjacent DIE. Layer VI is obliquely oriented, so that the long axes of the neurons are parallel to the surface of the medial bank of the rhinal fissure. Overall, both layers V and VI make an oblique and in some instances slightly curved border with the adjacent perirhinal cortex. 
Field DLE has previously not been defined as a separate division of rat entorhinal cortex, although Steward (1976) described a ’transitional’ field that corresponds to the presently defined field. In all other available descriptions, DLE is included into the lateral or the dorsolateral entorhinal cortex (Krettek and Price, 1977; Ruth et al., 1982).  
 

Entorhinal Cortex (EC)

Boundaries associated with this structure: 

Olfactory and periamygdaloid cortex
Postrhinal (POR)
Parasubiculum (PaS)
Perirhinal cortex, area 35 (PER35)


The entorhinal cortex (EC) occupies the ventro-caudal part of the cerebral hemisphere where it forms a cap-like structure. Its surface can be viewed as an ellipsoid with the white matter of the angular bundle as its center. The entorhinal cortex is hodologically defined by axonal projections from layer II neurons to the dentate gyrus of the hippocampal formation. Anteriorly, the entorhinal cortex is flanked by the piriform cortex laterally, and by the periamygdaloid cortex and the posterior cortical nucleus of the amygdala, medially (non-hippocampal regions not further described in this atlas). The transition between the entorhinal cortex and its anterior neighbors is approximately at the midst of the amygdaloid fissure, where the entorhinal cortex progressively decreases in width, such that it eventually extends anteriorly for approximately 2 mm as a narrow strip.

This anterior extension is delimited dorsolaterally by the perirhinal cortex and ventromedially by the piriform cortex. At its laterocaudal site, the entorhinal cortex is surrounded by the perirhinal and postrhinal cortices. Medially, the entorhinal cortex is bordered over most of its rostrocaudal extent by the parasubiculum. For further cytoarchitectonic descriptions, see its 5 subdivisions; dorsal-lateral entorhinal area (DLE), dorsal-intermediate entorhinal area (DIE), ventral-intermediate entorhinal area (VIE), medial entorhinal area (ME) and caudal entorhinal area (CE).

The lateral and caudal borders between EC and the perirhinal and postrhinal cortices are characterized by: 

1. the disappearance of the lamina dissecans 
2. the part of EC at the border stains densely for parvalbumin whereas the adjacent perirhinal and postrhinal cortex are almost devoid of positive staining 
3. the latter areas stain for calbindin which is much less conspicuously present in the adjacent parts of EC 
4. layer II of EC is characterized by a population of large to medium sized neurons that stain very densely for neuronal markers such as Nissl or NeuN. The adjacent parts of perirhinal and postrhinal cortices are characterized by blended layers II and III consisting of small, lightly stained neurons

The anterior border of EC is indicated by: 

1. a reduction in the number of cell layers from six down to three 
2. weak overall staining for parvalbumin in both anterior parts of EC as well as the neighboring areas, so this does not provide a clear indication of the border 
3. staining for calbindin that shows less dense staining in the anterior portions of EC compared to the adjacent regions 

The medial border with the parasubiculum features: 

1. a striking merge between layers II and III in the parasubiculum 
2. decreased staining intensity and homogeneity for calbindin in the neuropil of all layers 
3. dense staining for acetylcholinesterase in layers I-III in parasubiculum
 

Fasciola cinereum

Boundaries associated with this structure: 

(See text below)

The fasciolarum cinereum (FC), also referred to as gyrus fasciolaris, presents itself as a medially oriented small continuation of the most anterior part of the hippocampal formation. This is most easily appreciated in horizontal sections taken at dorsal levels. Although very little is known about FC and the borders with the remainder of the hippocampal formation are not that easy to discern, in some histological sections it can be distinguished rather easily. The FC can be further subdivided into a portion that contains granular cells similar to those found in the dentate gyrus (DG), which we will refer to as FC-DG. A second part mainly contains pyramidal cells, and this is further subdivided in a large-celled area that has similarities to CA3, referred to as FC-CA3 and a small-celled portion, a look-alike of CA1, named FC-CA1. In some sections these subdivisions are not easy differentiated, in which case we will use FC to indicate the entire complex. To understand the relationship of the FC with the remainder of the hippocampal formation, a through account of the embryological development is needed (see Stephan, 1975 for further details). Developmental data also clarify the relationship with two other presumed hippocampus-related structures in the brain, the induseum griseum and the anterior or dorsal tenia tecta. Neither one of those is included in the current account.

 
 

Medial entorhinal area (ME)

Boundaries associated with this structure: 
Caudal entorhinal field (CE)
Ventral-intermediate entorhinal field (VIE)


The medial entorhinal area occupies the most ventromedial portion of the entorhinal cortex. In most animals, the ME field is located on the medial bank of the amygdaloid fissure. Its overall cytoarchitectoninc features are rather similar to those of the caudal entorhinal area (CE), in that it shows a well defined laminar separation with a clear lamina dissecans, columnar organization in the deep layer V and VI, and an overall homogeneous cell density in layers II and III. 

Its caudal and caudomedial border is with CE, and its medial border is with the parasubiculum. Rostromedially, the parasubiculum is flanked by the most anterior remnant of the presubiculum. Laterally and anteriorly, it borders the ventral intermediate entorhinal field. A small portion of its rostromedial border is continuous with the periamygdaloid area. 

Together with the ventral-intermediate entorhinal area (VIE), field ME has several features in common with the rather ill-defined, so-called intermediate entorhinal cortex of various authors (Blackstad, 1956; Steward, 1976; Wyss, 1981; Ruth et al., 1982, 1988).
 

Parasubiculum (PaS)

Boundaries associated with this structure: 

Entorhinal cortex (EC)
Postrhinal cortex (POR)
Presubiculum (PrS)


The presubiculum (PrS) and the parasubiculum (PaS) are two multilayered cortical areas positioned in between the subiculum (Sub) laterally and the entorhinal cortex (EC) or retrosplenial cortex medially. Cytoarchitectonically, PrS and PaS are alike in that, with the exception of dorsal portions of PrS, the laminar differentiation is not very conspicuous. Similar to EC, both areas have their superficial layers II and III separated from the deep layers V and VI by a prominent cell free layer or lamina dissecans. The presubiculum superficial sheet of cells consists of darkly stained small pyramidal cells. The most superficial cells, forming layer II, are the most densely packed, while the more deeper cells in layer III show a more loose arrangement. The differentiation between layers II and III becomes clearer at more dorsal levels and the cells in layer II tend to be arranged in densely packed clusters dorsally. Layer V consists of one or two rows of large pyramidal cells, and layer VI harbors a variety of neuronal types. In contrast, neurons in layers II and III of PaS are fairly large and lightly stained, and the layers are generally not differentiated from one another. The deep layers are quite similar to those in PrS and EC.The PaS, or Brodman’s area 49 has been subdivided into two portions (area 49a and 49b) (Brodman, 1909) but the subdivisions are not applied in the present account.  
 

Perirhinal cortex (PER)

Boundaries associated with this structure: 

Insular cortex

Postrhinal cortex (POR)


The perirhinal cortex comprises two narrow strips of cortex, areas 35 and 36 (often referred to as perirhinal and ectorhinal cortex respectively; Paxinos & Watson, 1998), which are adjacent to one another and situated approximately along the third quarter of the rhinal sulcus. Anteriorly, the perirhinal cortex includes the fundus of the rhinal sulcus, both banks, and the dorsally adjacent cortex. Moving caudally, areas 35 and 36 are situated inside and dorsal to the fundus of the posterior rhinal sulcus. The agranular insular cortex (AIp) and granular/dysgranular insular cortex form the cortical areas located anterior to the perirhinal cortex. Area 36 is located caudal to the granular/dysgranular portion of the insular cortex, and area 35 is located caudal to AIp. It is generally accepted that when the claustrum is no longer visible, insular cortex is no longer present and is replaced by the perirhinal cortex (Burwell 1995).
 

Perirhinal cortex, area 35 (perirhinal; PER 35)

Boundaries associated with this structure: 

Entorhinal cortex (EC)
Perirhinal cortex, area 36 (ectorhinal; PER 36)

Area 35 is bordered dorsally by area 36 and ventrally and caudally by the entorhinal cortex. Just as area 36v may appear at a more anterior level than area 36d, area 35 may appear at a more anterior level than area 36. Area 35 is distinguished from the dorsally adjacent area 36 by several characteristics. First, layer I tends to be thicker, although this does not change sharply at the border between the two regions. The thickness of layer I in mid-dorsoventral area 36 is approximately 50% of that in mid-dorsoventral area 35. It may be, however, that the thickened layer I is a feature that is more appropriately associated with the rhinal sulcus than with a cytoarchitectural region. Second, the cells in area 35 exhibit a modified radial organization such that they form a shallow U-shaped arch beginning at the pial surface ventral to the rhinal sulcus and ending at the white matter deep to the rhinal sulcus. Third, area 35 is characterized by large, darkly stained, heart-shaped pyramidal cells in layer V. These cells are progressively smaller proceeding caudally. Although similar cells are seen in layer V of area 36v, there are fewer and they are not as distinctively heart shaped or as large. Heart-shaped cells become progressively smaller as one moves caudally.
 

Perirhinal cortex, area 36 (ectorhinal; PER 36)

Boundaries associated with this structure: 

Perirhinal cortex, area 35 (perirhinal; PER 35)
Temporal cortex
Insular cortex


Area 36 is bordered dorsally by the ventral temporal cortex (TeV), ventrally by area 35 and caudally by the postrhinal cortex . In general, area 36 is characterized by a patchy layer II composed of aggregates of lightly-stained, medium-sized cells. Layer II of TeV also appears patchy in some animals, especially at anterior levels, but can be distinguished from area 36 because the cells are more typically pyramid-shaped and usually more lightly stained.

In area 36, granular cells are apparent, but do not form a discrete layer rather, layer IV appears to merge with layer V. Area 36 is also characterized by a thick, bilaminate layer VI that distinguishes it both from the dorsally adjacent TeV and the ventrally adjacent area 35. The outer sublayer is similar to layer V in packing density and staining characteristics of cells. In the inner sublayer, cells are flattened parallel to the surface of the external capsule. This bilaminate layer VI, as well as the patchy layer II, are probably the features of area 36 that are seen most reliably across individual animals. Area 36 can be subdivided into three subfields: areas 36d, 36v, and 36p according to Burwell (1995).  

Postrhinal cortex

Boundaries associated with this structure: 

Entorhinal cortex (EC)
Parasubiculum (PaS)
Perirhinal cortex (PER)
Visual/temporal cortex


The postrhinal cortex is located caudal to area 36p and largely dorsal to the rhinal sulcus. In most cases, the postrhinal cortex arises at the caudal limit of the angular bundle when subicular cells are no longer present in coronal sections. Another landmark is the shortening of the presubiculum in the dorsoventral dimension and the imposition of a cell-sparse region deep to presubiculum that borders the underlying white matter. Like the perirhinal cortex, the postrhinal cortex is associated with the rhinal sulcus. Anteriorly, the superficial layers lie in the fundus of the rhinal sulcus, but the deep cortical layers underlying the fundus belong to the ventrally adjacent entorhinal cortex. Caudally, the region assumes a position above the fundus. 

If one imagines a caudal extension of the rhinal sulcus, it would rise at caudal levels and wrap around the caudal pole of the brain just ventral to the postrhinal cortex. If the cortex surrounding the rhinal sulcus and its imagined caudal extension could be straightened and flattened, the postrhinal cortex would form a long narrow strip largely dorsal to the sulcus and similar to the shape of the perirhinal cortex, but shorter along the longitudinal axis. The postrhinal cortex rises steeply and wraps obliquely around the caudal pole of the brain. Thus, its conformation is difficult to discern in the coronal plane. Because of the oblique cut in the coronal sections, the region extends farther dorsally and is limited in its rostrocaudal extent. Even unfolded maps can be misleading because of the tendency of surface areas of polar regions to be underrepresented. 

In sagittal sections, its long, narrow shape is more easily identified. The dorsal border of the postrhinal cortex is difficult to discern but is reliably identified relative to certain structural landmarks, particularly the location of the parasubiculum. The parasubiculum is on the medial cortical surface and is easily identified in cell-stained and acetylcholinesterase-stained sections. The dorsal border of POR on the lateral cortical surface is located directly across from the mid-dorsoventral level of the parasubiculum. Caudally, the parasubiculum also is useful in identifying the ventral border of the postrhinal cortex. At its caudal limit, the parasubiculum extends laterally more than halfway across the cortex and lying between the postrhinal cortex and entorhinal cortex. Thus, at this level, the postrhinal cortex has a modified triangular shape such that parasubiculum (medially) and the entorhinal cortex (laterally) form one side of the triangle and the pial surface and the lateral visual association cortex (VISl) form the other two sides. 

Perhaps the most characteristic cytoarchitectonic feature of the postrhinal cortex is its homogeneous packing density across layers II and IV and the resulting lack of a prominent laminar structure. It is difficult to differentiate deep from superficial layers because the layers appear to blend into one another. A second characteristic of the region visible in coronal sections is that all layers become unusually broad as one moves from superficial to deep. This thickening, however, is entirely a function of the conformation of the region and the plane of sectioning; the postrhinal cortex wraps around the caudal pole of the brain, and at these levels the coronal plane cuts obliquely across the radial axis of the cortex. A third characteristic of the postrhinal cortex is also due to conformation: the surface of the ventral portion of the region that is located dorsal to the rhinal sulcus is tightly convex such that deeper layers are compressed. Similar to cortical layers of gyri of the primate brain, the length of the superficial layers is longer than the length of the deeper layers. As a result, only a very narrow segment of layer VI is associated with the superficial layers of ventral postrhinal cortex. Although the lack of a prominent laminar structure is accentuated in coronal sections, this feature is also apparent in sagittal sections. The broadening of deeper layers, however, is not apparent in sagittal sections in which layers I, III, V, and VI occupy approximately one-third each of the radial extent of the cortex. The postrhinal cortex has two subfields: PORv and PORd. PORv is located dorsal to the entorhinal cortex and caudal to area 36p. 

In the coronal plane, in rare cases, PORv emerges anterior to PORd, and in these cases, PORv is located ventral to 36p at its most caudal levels. In most cases, however, area 35d occupies this position. PORd is located dorsal to PORv, but sometimes begins slightly more caudally than PORv. Cells in PORd layer III are more heterogeneous in size, shape, and color and are more organized and radial in appearance than in the ventrally adjacent PORv. Layers II and III are each composed of a homogeneous population of medium-sized, lightly stained round and polygonal cells, but the cells are more densely packed in layer II. In some cases, small dark pyramids are mixed into layer II. In the dorsally adjacent visual/temporal cortex (Tev), the cellular packing density is also higher in layer II; however, layer II and III cells of the dorsally adjacent Tev at this level are small, round, and darkly stained and do not have a radial appearance as in PORd. A granular layer is distinguishable, but less so at caudal levels. Layer V of PORd is slightly narrower than in PORv. Layer V differs from the dorsally located Tev in that Tev layer V is more open and sparsely populated, and the cells are larger. 

There are several typical cytoarchitectonic features of PORv. Perhaps the most distinctive is the presence of ectopic layer II cells at anterior levels of the region near the border with entorhinal cortex. These ectopic cells are present in all cases, but they vary in prominence. Layer II cells are moderately large, light, and round, but not as large as those seen in perirhinal cortex. Anteriorly, layers II and III can be distinguished from one another because layer III cells are less organized and less densely packed. Caudally, however, layer II is not easily distinguished from Layer III. PORv is dysgranular at all rostrocaudal levels, such that granule cells fill in between layers III and V, giving an overall homogeneous appearance to the region. Anteriorly, the width of layer V appears broader than in the dorsally located PORd, but this may be secondary to the curvature of the cortex at this level. Layer V is composed of small pyramid-shaped cells. Layer VI, which is fused together with layer V, is composed of fusiform cells and elongated pyramids that are oriented almost parallel with the angular bundle; however, only a small portion of layer VI is associated with PORv.  
 

Presubiculum (PrS)

Boundaries associated with this structure: 

Parasubiculum (PaS)
Subiculum (Sub)


The presubiculum (PrS) and the parasubiculum (PaS) are two multilayered cortical areas positioned in between the subiculum (Sub) laterally and the entorhinal cortex (EC) or retrosplenial cortex medially. Cytoarchitectonically, PrS and PaS are alike in that, with the exception of dorsal portions of PrS, the laminar differentiation is not very conspicuous. Similar to EC, both areas have their superficial layers II and III separated from the deep layers V and VI by a prominent cell free layer or lamina dissecans. The presubiculum superficial sheet of cells consists of darkly stained small pyramidal cells. The most superficial cells, forming layer II, are the most densely packed, while the more deeper cells in layer III show a more loose arrangement. The differentiation between layers II and III becomes clearer at more dorsal levels and the cells in layer II tend to be arranged in densely packed clusters dorsally. Layer V consists of one or two rows of large pyramidal cells, and layer VI harbors a variety of neuronal types. In contrast, neurons in layers II and III of PaS are fairly large and lightly stained, and the layers are generally not differentiated from one another. The deep layers are quite similar to those in PrS and EC.

The PrS, also referred to as Brodmans area 27 (Brodmann, 1909), is generally divided into a dorsal and a ventral portion, based on the different cytoarchitectonic features described above, which coincide with subtle differences in connectivity (van Groen, 1990). The dorsal portion has been named the postsubiculum, or Brodmans area 48 as well. The PaS, or Brodmans area 49 has been subdivided into two portions (area 49a and 49b) (Brodmann, 1909). Neither the Prs nor the PaS subdivisions are applied in the present account. 
 

Subiculum (SUB)

Boundaries associated with this structure: 

Cornu Ammonis 1 (CA1)
Presubiculum (PrS)

Area CA1 is part of the hippocampus proper, whereas the subiculum is generally considered to represent a separate entity. However, both areas share the layered composition which is typical for hippocampal fields. This layered apprearance is most easily seen in Nissl, NeuN or Timm-stained sections. In CA1, and less prominently in the subiculum, a wide molecular layer is located between the pyramidal layer and the hippocampal fissure. In the subiculum, this molecular layer can be subdivided into a deeper portion that is continuous with the stratum radiatum of CA1, and a superficial portion that is continuous with stratum lacunosum-moleculare. Note that this subdivision is difficult to detect by the markers used in this atlas. The superficial molecular layer of the subiculum and its continuation in stratum lacunosum moleculare of CA1 contain the perforant pathway fibers from the EC. Moreover, afferents from other structures such as the nucleus reuniens of the midline thalamus take a similar course and distribution.

CA1 can be easily recognised in Nissl and NeuN-stained sections due to its layer of neatly aligned pyramidal cells. This feature distinguishes CA1 from the subiculum, which dorsally is located between area CA1 and the retrospenial cortex, and ventrally between area CA1 and pre- and parasubiculum. The CA1/subiculum border is clearly marked by an abrupt widening of the pyramidal cell layer. Moreover, in material stained for parvalbumin or AChE, the pyramidal layer of CA1 is darkly stained, whereas the pyramidal cell layer of the subiculum is more diffusely stained, thus indicating a marked border between the two fields. In the TIMM staining the border is also visible, but then with an unstained CA1pyramidal layer, and a darkly stained cell layer in the proximal part of the subiculum. Although the border may appear easy to establish, it is in practice not possible to determine whether a particular dendrite at the border belongs to a neurone in CA1 or in the subiculum. This is caused by the oblique orientation of the CA1 / subiculum border relative to the transverse axis. Therefore, cells (in the stratum oriens, stratum radiatum, or lacunosum moleculare) close to this border do not necessary extent their dendrites perpendicularly to the orientation of stratum pyramidale. 
 

Ventral intermediate entorhinal area (VIE)

Boundaries associated with this structure: 

Caudal entorhinal field (CE)
Dorsal-intermediate entorhinal area (DIE)
Medial entorhinal field (ME)


Caudally, the ventral-intermediate entorhinal area (VIE) starts at the same level, or very closely to the medial entorhinal area (ME), which borders VIE medially. The anterior border of VIE is with the amygdalo-entorhinal transitional field (AE; see below) laterally, and with the peryamygdaloid cortex, medially. 

The layers of area VIE are better developed caudally than anteriorly, where they become quite condensed and tend to merge together. Layer I generally tends to be narrow. Layer II is also narrow and organized in clumps that are very close to each other, rendering a continuous appearance when studied with low magnification. Layer III is made up of densely packed, medium-size pyramids, although not as densely packed as in layer III of ME. Although very little or no separation exists between layers III and V, they can still be easily differentiated since layer V is characterized by the presence of big and dark pyramidal cells. Such big neurons are present not only in the outer part of the layer as in ME, but throughout the depth of layer V. 

The cell density of layer V decreases at the border with layer VI, which is rather thin and has a clear border with the underlying white matter. The main features of field VIE are comparable to those described by Haug (1976) for the caudomedial part of area 28L8 and the caudal parts of VLEA as described by Krettek and Price (1977). Together with ME, field VIE has several features in common with the rather ill-defined, so-called intermediate entorhinal cortex of various authors (Blackstad, 1956; Steward, 1976; Wyss, 1981; Ruth et al., 1982, 1988). 

Borders

Caudal entorhinal area (CE) / Dorsal intermediate entorhinal area (DIE)

Structures associated with this boundary: 
Caudal entorhinal area (CE) 
Dorsal intermediate entorhinal area (DIE)

The border of the caudal entorhinal area (CE) with the ventral-intermediate entorhinal area (VIE) & the dorsal-intermediate entorhinal area (DIE) is indicated by changes in the overall architecture in that the conspicuous lamina dissecans present in CE is less apparent in VIE & DIE. Layer III changes from having an overall regularly organization with equally spaced neurons in CE to a much more irregular layer in VIE & DIE where the neurons tend to be clustered with empty areas in between. The columnar like arrangements seen in layers V and VI of CE also breaks down in VIE & DIE such that the appearance of layers V and VI is more laminar. Layer II of VIE is much thinner that in CE, whereas in CE and DIE this layer has an overall comparable appearance. When comparing sections stained for parvalbumin, it is obvious that CE has very strong positivity contrasting with the moderate to weak staining in DIE and VIE respectively. Stainings for the presence of calbindin mark the border since in CE positivity is almost exclusive for layer II whereas in VIE & DIE staining is also present in layers I and III. 


Caudal entorhinal area (CE) / Dorsolateral entorhinal area (DLE)

Structures associated with this boundary: 

Caudal entorhinal area (CE)
Dorsolateral entorhinal area (DLE)

At some levels the caudal entorhinal area (CE) meets the dorsal-lateral entorhinal area (DLE), which can be quite confusing when looking at coronal sections. Both areas have large darkly stained layer II neurons and the only subtle difference is that the layer II cells in DLE tend to spread into the molecular layer, a feature not seen in CE. In addition, the deep and superficial delineation of the lamina dissecans is much more clear in CE than in DLE. Immunoreactivity for calbindin provides valuable additional information since CE shows strong staining almost exclusively in layer II, whereas in calbinding positive neuropil is also present in layer III of DLE. 


Caudal entorhinal field (CE) / Medial entorhinal field (ME)

Structures associated with this boundary: 

Caudal entorhinal field (CE)
Medial entorhinal field (ME)

This border is indicated by a marked change in the architecture of layer II. The rather homogeneous layer II of the caudal entorhinal area (CE) is not present in the medial entorhinal area (ME), where this layer generally is broken up into two or three clusters of cells. In ME, layer II is separated from layer III by a narrow, irregular cell sparse zone, and layer III is less regular than in CE. Quite often layer III of ME can be subdivided into an outer more densely packed zone, and an inner less dense zone. Layer V in both areas can be subdivided into sublayers, of which the superficial layer Va in ME contains a population of large, darkly stained pyramidal cells, not present in CE. When stained for the presence of calbindin, CE shows strong staining of both neuropil and neurons almost exclusively in layer II, whereas in ME this staining extends into the superficial portion of layer III. The border is also signaled by a change in the staining for parvalbumin. In CE layers II and III exhibit moderate homogeneous staining whereas in ME, parvalbumin positive neuropil is rather sparse.  

Caudal entorhinal area (CE) / Ventral Intermediate Entorhinal Cortex (VIE)

Structures associated with this boundary: 
Caudal entorhinal area (CE)
Ventral Intermediate Entorhinal Cortex (VIE)

The border of the caudal entorhinal area (CE) with the ventral-intermediate entorhinal area (VIE) & the dorsal-intermediate entorhinal area (DIE) is indicated by changes in the overall architecture in that the conspicuous lamina dissecans present in CE is less apparent in VIE & DIE. Layer III changes from having an overall regularly organization with equally spaced neurons in CE to a much more irregular layer in VIE & DIE where the neurons tend to be clustered with empty areas in between. The columnar like arrangements seen in layers V and VI of CE also breaks down in VIE & DIE such that the appearance of layers V and VI is more laminar. Layer II of VIE is much thinner that in CE, whereas in CE and DIE this layer has an overall comparable appearance. When comparing sections stained for parvalbumin, it is obvious that CE has very strong positivity contrasting with the moderate to weak staining in DIE and VIE respectively. Stainings for the presence of calbindin mark the border since in CE positivity is almost exclusive for layer II whereas in VIE & DIE staining is also present in layers I and III. 


CA1 / CA2

Structures associated with this boundary: 

CA1
CA2

The border between CA2 and CA1 coincides with a sudden drop in the size of neurons in the principal cell layer in CA1, which can be observed in sections stained for NeuN. In addition, in CA1, pyramidal cells in the most superficial layer stain positive for calbindin and there is a salient calbindin positive band in the neuropil, at the border between stratum radiatum and stratum lacunosum-moleculare, this band is completely absent in CA2. Finally, in CA1, the neuropil of stratum lacunosum-moleculare stains positive for parvalbumin, which is not seen in CA2. 


CA1 / Subiculum (SUB)

Structures associated with this boundary:  

CA1
Subiculum (SUB)

At the border between CA1 and Sub, the condensed cell layer of rather homogeneously looking cell bodies of CA1 changes into a less dense, multilayered structure as seen in sections stained for neuronal markers such as NeuN. There is an abrupt change in the staining for calbindin in that the positive cells and rather intense stained neuropil of stratum pyramidale of CA1, and also the staining in stratum lacunosum-moleculare, is replaced by the almost complete negative corresponding layer in Sub. This change coincides with a change in the appearance of the superficial fiber layer. In CA1, one can rather easily differentiate between stratum radiatum, characterized by the aligned orientation of apical dendrites and the more superficial stratum lacunosum-moleculare. This laminar separation is absent in Sub. In sections stained for parvalbumin, the pyramidal layer of CA1 is darkly stained, whereas the pyramidal cell layer of the subiculum is more diffusely, but still rather densely stained, thus indicating a marked border between the two fields. 


CA2 / CA3

Structures associated with this boundary: 
CA2
CA3

These two parts of the Ammon’s horn are characterized by presence of large pyramidal cells, that form a few closely packed layers on top of one another. Both areas have a layered appearance, similar to CA1 and the subiculum. In the rat the following features differentiate CA3 clearly from CA2: The pyramidal cell layer of CA2 tends to be slightly thicker than in the adjacent CA3, and is populated by a mixture of large cells, similar in size as the CA3 neurons, and small neurons, similar in size to CA1 neurons. CA3 is characterized by the strong calbindin positive mossy fiber input, which is almost completely absent in CA2. Using a stain for parvalbumin, it is obvious that CA2 contains a much higher density of fairly large positive neurons than CA3.  

 

CA3 / Dentate Gyrus (DG)

Structures associated with this boundary: 
CA3
Dentate gyrus (DG) 

The dentate gyrus is characterized by the presence of a curved cell layer, densely packed with granule cells, i.e. cells that lacks basal dendrites but have an extensive apical dendritic tuft, extending into the molecular layer. The main border to be established is between the hilus of DG and the most proximal tip of CA3. Although this border may be difficult to observe in conventionally stained sections, it is distinguished by a rather abrupt change in neuropil staining for calbindin, such that the hilus is clearly positive for calbindin, different from the adjacent proximal part of CA3. This is caused by the calbindin-positive mossy fibers from the DG that reach cells in the hilus and the CA3. Likewise, the molecular layer of the DG stains strongly for calbindin, in contrast to the molecular layer of CA3. Note that this staining pattern is mimicked by staining with an antibody against dynorphin and using the Timm stain, not used in the series prepared for this atlas. 


Dorsal intermediate entorhinal area (DIE) / Dorsolateral entorhinal area (DLE)

Structures associated with this boundary: 
Dorsal intermediate entorhinal area (DIE)
Dorsolateral entorhinal area (DLE)

Although the overall cellular composition is quite similar in the dorsal-intermediate entorhinal area (DIE) and the dorsal-lateral entorhinal area (DLE), there are striking and defining differences in the way cells are organized into layers. Layer I of the dorsal-lateral entorhinal area (DLE) is thinner than that of the dorsal-intermediate entorhinal area (DIE). In both areas, layer II contains rather big, rounded neurons that stand as darkly stained in Nissl stained material. However, in DIE these cells are organized in a fairly dense and homogeneously packed layer, while they in DLE tend to be more dispersed and to some extent invade the molecular layer. 

A very narrow, relatively acellular band separates layer II from layer III in much of the DIE extent, which is not seen in DLE. Layer III in DIE is wide, and clearly presents a narrow, more densely packed outer zone with its neurons arranged in clusters, and a less densely and irregularly packed inner zone. In contrast, layer III of DLE is rather thin, and the cells are organized in horizontal rows, parallel to the surface curvature of the rhinal fissure. Area DIE does not have a distinguished layer IV or lamina dissecans whereas in DLE layers III and V are markedly separated from each other. Layers V and VI do not show striking architectonic differences. When considering additional markers it is apparent that whereas layer II of DLE stains rather strongly for calbindin both in terms of neurons as well as neuropil, in DIE the neuropil staining is less dense such that the individual positive neurons are easy to see. In material stained for parvalbumin, the DIE / DLE border coincides with a gradual loss of the positive staining in layer III: moderate in DLE, light to absent in DIE. 

Dorsal intermediate entorhinal area (DIE) / Ventral Intermediate Entorhinal Cortex (VIE)

Structures associated with this boundary: 
Dorsal intermediate entorhinal area (DIE)
Ventral Intermediate Entorhinal Cortex (VIE)

Layer II of the ventral-intermediate entorhinal area (VIE) is narrow and organized in clumps that are very close to each other, rendering a continuous appearance when studied with low magnification. In contrast, in the dorsal-intermediate entorhinal area (DIE), layer II contains rather big, rounded neurons that stand out in regular Nissl and NeuN stains since they are darkly stained. A very narrow, relatively acellular band separates layer II from layer III in much of the DIE extent. Layer III of VIE is made up of densely packed, medium-size pyramids, whereas in DIE this layer is wide, and clearly presents a narrow, more densely packed outer zone with its neurons arranged in cell clusters, and a less densely and irregularly packed inner zone. No major differences between the two areas are apparent in deep layers V and VI, with the exception that the border between layer V and the lamina dissecans is very much obscured in DIE compared to VIE and the latter area also contains much more large, darkly stained pyramidal cells than DIE does. 

Calbindin is strongly expressed in deep layer I, layer II and III in VIE, both in neurons and neuropil. In DIE, the staining in layers I and III seems to be less dense, but the staining of individual neurons in layer II is more intense in DIE than in VIE. Whereas we observed only very weak staining in VIE for parvalbumin. DIE shows an overall higher intensity when stained for parvalbumin, particularly in layers II and III. 

Entorhinal cortex (EC) / perirhinal cortex, area 35 (PER35)

Structures associated with this border: 
Entorhinal cortex
Perirhinal cortex, area 35 (PER35)

The border between EC and PER35 can be easily established based on cytoarchitectonic criteria, because in PER35 no lamina dissecans can be distinguished, a distinct feature of EC. Moreover, layers II and III of PER35 are characterized by the presence of lightly stained densely packed neurons. This results in an appearance that is quite different from the adjacent EC, where cells in layer II are larger and stain darker for Nissl and NeuN substance. The laminar differentiation in EC layers II and III is much more distinct. Using an antibody against the calcium-binding protein parvalbumin, the border between EC and PER35 of the perirhinal cortex stands out. While, the entorhinal cortex close to the border stains heavily positive for parvalbumin, an abrupt loss of staining is seen in the neighboring area 35. By contrast, a marked increase of staining for the calcium binding protein calbindin is noticeable in area 35. 


Entorhinal Cortex / Olfactory and Periamygdaloid Cortex

Structures associated with this boundary: 
Entorhinal Cortex
Olfactory and periamygdaloid cortex

The anterior border of EC is indicated by a reduction in the number of cell layers from six (molecular layer, cell layers II, III, V, and VI, and the lamina dissecans) to three (molecular layer, principal cell layer (layer II) and polymorph layer (layer III)). While this border is not easily identified in sections stained for parvalbumin, the anterior portions of EC stain less densely for calbindin compared to the neighboring olfactory and periamygdaloid areas.


Entorhinal Cortex / Parasubiculum

Structures associated with this boundary: 
Entorhinal cortex
Parasubiculum

The medial border of EC with PaS features a striking merging of layers II and III in PaS, so that the layers II and III are difficult to distinguish here. The darkly stained cells in layer II of the EC tend to be clustered (as seen in sections stained for NeuN), and the deepest cells are strongly positive for calbindin. In the ventral part of PaS, an overall homogeneous but weak staining for calbindin is present superficially, while sparsely distributed cells stained for calbindin are seen in layers III and V. Although not shown in this atlas, the EC / PaS boundary coincides with a marked increase in staining for acetylcholinesterase in layers I-III in parasubiculum. Staining for parvalbumin is not a distinguishing feature for this border. 

Entorhinal Cortex / Postrhinal Cortex

Structures associated with this boundary: 
Entorhinal cortex
Postrhinal cortex

The border between EC and POR is rather easy to establish based on cytoarchitecture, because in POR no lamina dissecans can be distinguished. Moreover, layers II -IV of POR are characterized by their homogeneous packing density of relatively small, lightly stained neurons, resulting in an overall lack of a clear laminar structure. This is quite different in the adjacent EC where layer II cells are significantly larger and darker in Nissl and NeuN staining. Also layers III and V of the EC have an overall different appearance.
A further distinctive characteristic of POR is the presence of ectopic layer II cells in the more rostral portions of POR. Staining with an antibody against the calcium-binding protein parvalbumin makes the border between EC and POR stands out. The EC, however, stains strongly positive for parvalbumin close to the border, so that an abrupt loss of staining can be seen in the POR. By contrast, in material stained for the calcium-binding protein calbindin, a marked, strong staining in POR is noticeable.


Medial Entorhinal Area (ME) / Ventral Intermediate Entorhinal Cortex (VIE)

Structures associated with this boundary: 
Medial Entorhinal Area (ME)
Ventral Intermediate Entorhinal Cortex (VIE)

Layer II of the ventral-intermediate entorhinal area (VIE) is narrow and organized in clumps that are very close to each other, rendering a continuous appearance when studied with low magnification. In the medial entorhinal area (ME), layer II is thicker and more continuous. Layer III of both areas is made up of medium-size pyramids, which in VIE are not as densely packed as in layer III of ME. Although very little or no separation exists between layers III and V in VIE, they can still be differentiated since layer V is characterized by the presence of darkly stained, big pyramidal cells. In ME, these large pyramidal neurons are only present in the outer part of layer V, while they are present throughout the depth of layer V in VIE.When stained for calbindin, the VIE / ME border stands out: in ME, calbindin-positive neuropil is found mainly in layers I, II and superficial III, while in VIE the staining is much weaker in the superficial half of layer I, while extending throughout layer III. In addition, the density of stained neurons is much higher in layer II of ME compared to VIE. Both areas are characterized by virtual absence of parvalbumin positivity. 


Parasubiculum (PaS) / Postrhinal cortex

Structures associated with this boundary: 
Parasubiculum (PaS)
Postrhinal cortex

The border between the postrhinal cortex (POR) and the parasubiculum (PaS) can easily be established using cytoarchitectonic criteria since the overall cell size in layers II and III of both areas is so different with PaS having the larger cells (see NeuN stain). However, due to the very complicated intermingling of the two areas, in particular when studying coronal or horizontal sections, it is quite difficult to establish borders with certainty. Sections stained for the presence of parvalbumin and calbindin make the borders stand out clearly, irrespective of the plane of sectioning. Whereas PaS stains densely for parvalbumin, POR does not. Using calbindin as a marker, the superficial layers of POR are positive, whereas in PaS only a few neurons superficially in layer V are densely stained and the remaining layers are almost completely devoid of labeling. 


Parasubiculum (PaS) / Presubiculum (PrS)

Structures associated with this boundary: 
Parasubiculum (PaS)
Presubiculum (PrS)


Cytoarchitectonically, the presubiculum (PrS) and the parasubiculum (PaS) are alike in that, with the exception of dorsal portions of PrS, the laminar differentiation between the two layers is not very conspicuous. However, the distinctive differentiation is the cell size. Whereas cells in PaS are fairly large, almost the same size of layer II cells in EC, those in PrS are much smaller. In addition, the latter also stain darker in a Nissl or NeuN stain.

The border in the deep layers V and VI however is impossible to establish, resulting in the widely accepted custom not to differentiate between the deep layers at all (similarly, no clear border is generally indicated between the deep layers of PaS and EC). However, using the additional information provided by staining with an antibody against calbindin and parvalbumin, a marked border between the two regions can be established. PaS has almost no neuropil staining for calbindin, whereas PrS does show strong staining in neurons and neuropil in layer II. Parvalbumin staining provides a corresponding border since layers II and III of PaS stain very dense and almost homogeneously whereas the deep layers are almost negative. In PrS, staining is dense in layer II, somewhat weaker in layer III and in addition the superficial portion of layer V shows positive neuropil with an occasional positive neuron, whereas the remainder of the deep layers show a slightly less dense staining. 

The PrS, also referred to as Brodman’s area 27 (Brodmann, 1909), is generally divided into a dorsal and a ventral portion, based on the different cytoarchitectonic features described above, which coincide with subtle differences in connectivity (van Groen and Wyss, 1990). The dorsal portion has been named the postsubiculum, or Brodman’s area 48 as well. The PaS, or Brodman’s area 49 has been subdivided into two portions (area 49a and 49b,(Brodman, 1909)). Neither the PrS nor the PaS subdivisions are applied in the present account. 


Insular Cortex / Perirhinal Cortex

Structures associated with this boundary: 
Insular Cortex
Perirhinal Cortex

The border between the perirhinal cortex (PER) and insular cortex (Ins) coincides with the disappearance of the claustrum, a structure that underlies the insular cortex. This structure is absent at the level of the perirhinal cortex. The insular cortex has a clearly laminated appearance because layers II and III and layer VI appear darker with a lighter layer V in between. Layer V of the insular cortex is sandwiched between cell sparse layers, a feature not seen in PER. The insular cortex has a more or less well defined granular cell layer IV which is more conspicuous at more dorsal portions, whereas in the perirhinal cortex only area 36 (PER36) shows layer IV. In the insular cortex, layers V and VI can be easily differentiated from each other and they are of approximate similar thickness, both features are not obvious in either perirhinal areas 35 or 36. No apparent changes are present between PER and the insular cortex in staining patterns for parvalbumin and calbindin. 

Perirhinal (PER) / Postrhinal cortex (POR)

Structures associated with this boundary: 

Perirhinal cortex (PrS)
Postrhinal cortex (POR)

The main distinctive feature between perirhinal cortex (PER) and postrhinal cortex (POR) is the distinctive presence of ectopic cells in anterior portions of POR, which are absent in PER. In addition, cells in layer II of POR are somewhat smaller than in PER. Area 35 (PER35) lacks a clear laminar differentiation between layers II and III, whereas the anterior portion of POR does show a difference between the two layers. Finally, in POR an apparent granular layer IV can be distinguished, which is incomplete ventrally (dysgranular) and complete at more dorsal levels (granular). In contrast, PER35 is agranular. Staining for parvalbumin or calbindin does not reveal striking differences between PER and POR. 

Presubiculum (PrS) / Postrhinal cortex

Structures associated with this boundary: 
Presubiculum (PrS)
Postrhinal cortex   

Perirhinal cortex, area 35 (perirhinal; PER 35) / Perirhinal cortex, area 36 (ectorhinal; PER 36)

Structures associated with this boundary: 
Perirhinal cortex, area 35 (perirhinal; PER 35)
Perirhinal cortex, area 36 (ectorhinal; PER 36)

The border between perirhinal cortex area 35 nd 36 is characterized by an emerging differentiation between layers II and III in PER36 which is absent in PER35. Cells in layer II of PER36 are slightly more intensely stained and form a regularly packed dense cell layer. Cells in layer III of PER36 show a slightly less regular pattern compared to PER35. Layer V of PER35 is characterized by fairly large heart shaped pyramidal cells that in PER36 are much less abundant. Finally, whereas PER35 is agranular, i.e. lacks a clear granular layer IV, PER36 does express a layer IV although this is only well developed at more dorsal levels of PER36 such that at the border with PER35 this feature is not particularly obvious. Additional immunostains for parvalbumin and calbindin do not provide additional information to recognize this border. 


Perirhinal cortex, area 36 (ectorhinal; PER 36) / Insular Cortex

Structures associated with this boundary: 
Perirhinal cortex, area 36 
Insular Cortex

In lower magnification photomicrographs, insular cortex can be identified by its three-layered appearance. The cellular layers divide into thirds such that the superficial layers (II and IV) and deep layer (VI) appear darker than the intervening layer V, which has cell-sparse gaps on either side. This trilaminar look is not apparent in the caudally adjacent perirhinal cortex rather, there are no cell-sparse gaps and cells of layer V are more densely packed, especially in area 35. Several cytoarchitectonic details also distinguish the anteriorly adjacent insular areas from the perirhinal cortex area 36. In the insular cortex, layers V and VI are approximately of the same thickness and readily distinguished. Layer V is broad and composed of medium to large, darkly stained pyramidal cells. It is continuous across the two regions, forming a broad, homogeneous band with cells of similar size, shape, staining characteristics, and packing density. When the perirhinal cortex begins, the single homogeneous, band-like layer V is no longer present, but replaced bysmaller and more densely packed layer V cells. 

Perirhinal cortex, area 36 (ectorhinal; PER 36) / Temporal cortex (Te)

Structures associated with this boundary: 
Perirhinal cortex, area 36 (ectorhinal; PER 36)
Temporal cortex

The border between perirhinal cortex area 36 (PER36) and the temporal cortex (Te) is not easily established from cytoarchitectonic features, and even histochemical criteria do not help much. In general, layer II of Te can be distinguished from PER36 because the cells in Te are more typically pyramid-shaped and usually more lightly stained; quite often layer II in Te is also thinner compared to layer II of PER36, particularly at more posterior levels. In addition, the overall appearance of layer V differs since in PER36 there is no cell-sparse gap on either side of layer V, as is sometimes observed in the dorsally located Te. Also layers V and VI of PER36 are narrower than in area Te. With an immunostain for parvalbumin, area Te expresses increased staining intensity when compared to PER36. This increased expression of parvalbumin reactivity coincides with a change in staining for calbindin: in PER36 this staining is present almost homogeneously throughout all layers whereas Te expresses a more distinct laminated pattern. 

Postrhinal Cortex / Visual/Temporal Cortex

Structures associated with this boundary: 
Postrhinal cortex (POR)
Visual/Temporal Cortex

On the basis of cytoarchitectonic criteria, the dorsal border of the postrhinal cortex (POR) is difficult to establish throughout its extent. POR lacks a clear laminar structure which is more conspicous in the adjacent cortex. Layer II of temporal and primary visual cortex is more densely packed, and cells in layers II and III tend to be slightly smaller than in POR. Layer V in POR also is less dense compared to adjacent cortex and layer IV is less conspicuous in POR. Ectopic layer II cells, a distinct feature of POR are absent in the dorsally adjacent cortices. Using immunostaining for calbindin does not show much of a border although at the dorsal border of POR, there is a slight increase in staining intensity in deeper layers, resulting in a more bilaminar staining pattern. Sections stained for parvalbumin though provide a pretty strict border between since the dorsally adjacent cortex does stain with this marker whereas all layers of POR are almost completely negative except for the presence of an occasional positive neuron. 

Presubiculum (PrS) / Subiculum (SUB)

Structures associated with this boundary: 
Presubiculum (PrS)
Subiculum (SUB)

This border is fairly easy to establish, since in sections stained for a neuronal marker such as NeuN, there appears to be a sudden addition of a new cell layer on top of the continuation of the subiculum, leaving a cell sparse layer or lamina dissecans in between. In order to reliably position the border in the deep layers one has to combine information on cell density and cell size with the differential staining for parvalbumin. In SUB, most cells are approximately of the same size, with no apparent layered features being apparent. In PrS, the deep layers tend to show an indication of a differentiation between layers V and VI. Moreover, the intense positive staining for parvalbumin of the neuropil in the pyramidal cell layer of SUB is replaces by a much less conspicuous staining in layers V and VI of PrS. 

</div></div></div></div></div></div>