TIME OF ORIGIN OF BASAL FOREBRAIN NEURONS IN THE MOUSE: AN AUTORADIOGRAPHIC STUDY

Item Type text; Dissertation-Reproduction (electronic)

Authors Creps, Elaine Sue, 1946-

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 07/10/2021 05:12:20

Link to Item http://hdl.handle.net/10150/290321 INFORMATION TO USERS

This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted.

The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction.

1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Paga(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity.

2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame.

3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until complete.

4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced.

5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received.

Xerox University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48106 73-20,651

CREPS, Elaine Sue Edmonds, 1946- TIME OF ORIGIN OF BASAL FOREBRAIN NEURONS IN THE MOUSE: AN AUTORADIOGRAPHIC STUDY.

The University of Arizona, Ph.D., 1973 Anatomy

University Microfilms, A XEROX Company , Ann Arbor, Michigan

THIS DISSERTATION HAS BEEN MICROFLIMED EXACTLY AS RECEIVED TIME OF ORIGIN OF BASAL FOREBRAIN NEURONS

IN THE MOUSE: AN AUTORADIOGRAPHIC STUDY

by

Elaine Sue Creps

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF ANATOMY

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

19 7 3 THE UNIVERSITY OF ARIZONA

GRADUATE COLLEGE

I hereby recommend that this dissertation prepared under my direction by Elaine Sue C'reps entitled Time of Origin of Basal Forebrain Neurons

in the Mouse: An Autoradiographic Study be accepted as fulfilling the dissertation requirement of the degree of Doctor of Philosophy

, 7yUweA. , 1 cl 7 3 Dissertation Director Date

After inspection of the final copy of the dissertation, the follov/ing members of the Final Examination Committee concur in its approval and recommend its acceptance:-

/'• Ztv '#/ j& ••j- / y f /97J

fV.Z&aJt, "Tfadil /9?3

^Uu-,/

2. "y,

This approval and acceptance is contingent on the candidate's adequate performance and defense of this dissertation at the final oral examination. The inclusion of this sheet bound into the library copy of the dissertation is evidence of satisfactory performance at the final examination. STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduc­ tion of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his judgment the proposed use of the material is in the interests of schol­ arship. In all other instances, however, permission must be obtained from the author.

SIGNED: s&tfrur ACKNOWLEDGMENTS

This dissertation was made possible by its director, Dr. Jay

B. Angevine, Jr., to whom I will always be grateful. His support and interest, as well as his constant attention to the details of this manuscript, have turned an idea into a finished product. I would also like to express my appreciation to all members of the

Department of Anatomy who have given me assistance throughout my graduate training. I am especially grateful to Dr. John W. Sechrist for his patience, and many helpful suggestions during preparation of this paper. Thanks also are extended to Dr. Elizabeth Taber

Pierce, who generously loaned part of her slide collection for this study. I am indebted to Miss Almera Seber for her technical assistance and Michael Barnard for her technical advice. Special thanks also go to Mrs. Barbara Spalding for her careful typing of this manuscript.

iii TABLE OF CONTENTS

Page

LIST OF ILLUSTRATIONS vii

LIST OF TABLES xi

ABSTRACT xii

1. INTRODUCTION 1

2. HISTORICAL PERSPECTIVE 5

Proliferation in the Neural Tube ...... 5 Classicial Studies 5 Autoradiographic Studies 9 Recent Golgi and Electron Microscopic Studies 14 Differentiation of Neurons 17 Classical Studies 17 Electron Microscopic Studies 19 Recent Golgi Studies 22

3. AUTORADIOGRAPHIC STUDIES OF REGIONAL HISTOGENESIS 25

Spinal Cord 25 Myelencephalon 25 Metencephalon 28 Pons 28 Cerebellum 29 Mesencephalon 36 Diencephalon 37 Telencephalon 1+1 Olfactory Centers 4l Hippocampal Region 42 Cerebral Cortex 45 Other Telencephalic Centers 46

4. THE BASAL FOHEBRAIN 48

Olfactory Tracts and Anterior Commissure 49 Medial Forebrain Bundle 51 Fornix 52

iv V

TABLE OF CONTENTS--Continued

Page

Stria Terminalis 53 Diagonal Band of Broca 53 Stria Medullaris Thalami . 5^-

5. AUTORADIOGRAPHY 56

History 56 The Radioisotope 59 The Autoradiographic Process 60

6. MATERIALS AMD METHODS 62

Animals 62 Injection of Tritiated Thymidine 61+ Fixation . . 65 Embedding and Sectioning 66 Staining for Myelin 67 Autoradiography - Dipping ..... 67 Autoradiography - Development . 68 Nuclear Staining 69 Analysis of Autoradiograms 69

7. RESULTS 72

Basal Olfactory Region 73 Anterior Olfactory Nucleus 73 Nucleus of the Lateral Olfactory Tract 75 Olfactory Tubercle 75 Basal Islands 76 Preoptic Area 78 Periventricular Region 79 Medial Preoptic Nucleus . . 80 Lateral Preoptic Region 8l Magnocellular Preoptic Nucleus . . . . 82 Septal Area 82 Medial Septal Nucleus 84- Triangular Septal Nucleus «... 8U Fimbrial Septal Nucleus 85 Lateral Septal Nucleus 00... 85 Nucleus Accumbens Septi 86 Nucleus of the Diagonal Band of Broca 87 Interstitial Nucleus of the Stria Terminalis 88 Nucleus of the Anterior Commissure 89 Summary of Observations 89 vi

TABLE OF CONTENTS--Continued

Page

8. DISCUSSION 92

Significance of Results 92 Comparison of Results with Other Studies 92 Postnatal Neurogenesis 97 Gradients 100 Time of Origin Related to Other Events in Neurogenesis . . 105 Nucleus Accurabens - Septal or Striatal? 109 Phylogenetic Parallels of Time of Origin 110 The Question of the Telodiencephalic Boundary 112

APPENDIX: AUTORADIOGRAPHIC RECONSTRUCTIONS AND GRAPHS .... 117

LITERATURE CITED 175 LIST OF ILLUSTRATIONS

Figure Page

1. Basal and sagittal views of the mouse brain illustrating levels I-V used to prepare autoradiographic reconstructions 119

2. Pattern of labelling in anterior olfactory nucleus on Ell and early on E12 120

3. Pattern of labelling in anterior olfactory nucleus late on E12 and on E13 121

Pattern of labelling in anterior olfactory nucleus early on El^+ and late on El^- 122

5- Pattern of labelling in anterior olfactory nucleus on E15 and early on El6 123

6. Pattern of labelling in anterior olfactory nucleus on E17 and early on El8 12^4-

7. Pattern of labelling in anterior olfactory nucleus late on El8 and on postnatal day one . . . . 125

8. Pattern of labelling at rostral septal level on -ElO 126

9. Pattern of labelling at rostral septal level on Ell 127

10. Pattern of labelling at rostral septal level early on E12 128

11. Pattern of labelling at rostral septal level late on E12 129

12. Pattern of labelling at rostral septal level on E13 ^3°

13'. Pattern of labelling at rostral septal level early on El4 131

lb. Pattern of labelling at rostral septal level late on El^ 132

15• Pattern of labelling at rostral septal level on E15 ^-33

vii viii

LIST OF ILLUSTRATIONS--Continued

Figure Page

16. Pattern of labelling at rostral septal level early on El6 13^

17. Pattern of labelling at rostral septal level late on El6 135

18. Pattern of labelling at rostral septal level on E17 136

19. Pattern of labelling at rostral septal level early on El8 137

20. Pattern of labelling at rostral septal level late on El8 138

21. Pattern of labelling at rostral septal level on postnatal day one 139

22. Pattern of labelling at rostral septal level on postnatal day four lUo

23. Pattern of labelling at middle septal and rostral preoptic level on E10 lUl

2b. Pattern of labelling at middle septal and rostral preoptic level on Ell 1^2

25. Pattern of labelling at middle septal and rostral preoptic level early on E12 1^3

26. Pattern of labelling at middle septal and rostral preoptic level late on E12 lM+

27. Pattern of labelling at middle septal and rostral preoptic level on E13 1^5

28. Pattern of labelling at middle septal and rostral preoptic level early on El^f 1^+6

29. Pattern of labelling at middle septal and rostral preoptic level late on ElU IU7

30. Pattern of labelling at middle septal and rostral preoptic level on E15 1^8 ix

LIST OF ILLUSTHATIONS--Continued

Figure Page

31. Pattern of labelling at middle septal and rostral preoptic level early on El6 IU9

32. Pattern of labelling at middle septal and rostral preoptic level late on El6 150

33. Pattern of labelling at middle septal and rostral preoptic level on E17 151

3^. Pattern of labelling at caudal septal and preoptic level on E10 152

35• Pattern of labelling at caudal septal and preoptic level on Ell 153

36. Pattern of labelling at caudal septal and preoptic level early on E12 15^

37. Pattern of labelling at caudal septal and preoptic level late on E12 155

38. Pattern of labelling at caudal septal and preoptic level on E13 156

39- Pattern of labelling at caudal septal and preoptic level early on El^ 157

1+0. Pattern of labelling at caudal septal and preoptic level late on El^- 158

Ul. Pattern of labelling at caudal septal and preoptic level on E15 159

k2. Pattern of labelling at caudal septal and preoptic level early on El6 . . . • 160

U3. Pattern of labelling at caudal septal and preoptic level late on El6 1^1

Ml. Pattern of labelling at caudal septal and preoptic level on E17 162

1+5. Pattern of labelling at caudal septal and preoptic level early on El8 1&3 X

LIST OF ILLUSTRATIONS--Continued

Figure Page

k6. Pattern of labelling in the nucleus of the lateral olfactory tract on E10, Ell, early E12, E13, early El4, and late El4 l6U

kj. Number of heavily labelled neurons on each day of development in the anterior olfactory nucleus, pars externa and pars lateralis 165

U8. Number of heavily labelled neurons on each day of development in the anterior olfactory nucleus, pars medialis, pars dorsalis, and pars posterior .... l66

49. Number of heavily labelled neurons on each day of development in nucleus of the lateral olfactory tract, and olfactory tubercle, polymorph layer an pyramidal layer ...» 167

50. Number of heavily labelled neurons on each day of development in islands of Calleja, insula magna, and nucleus of the anterior commissure 168

51. Number of heavily labelled neurons on each day of development in median, periventricular, and suprachiasmatic preoptic nuclei 169

52. Number of heavily labelled neurons on each day of development in lateral and medial preoptic nuclei .... 170

53. Number of heavily labelled neurons on each day of development in magnocellular preoptic nucleus and nucleus of the diagonal band of Broca 171

54. Number of heavily labelled neurons on each day of development in medial and lateral septal nuclei 172

55^ Number of heavily labelled neurons on each day of development in triangular and fimbrial septal nuclei . . 173

56. Number of heavily labelled neurons on each day of development in interstitial nucleus of the stria terminalis and nucleus accumbens septi 17^ LIST OF TABLES

Table Page

1. Animals Employed in the Present Study 63

2. Comparison of Results with Other Studies . 93

3. List of Abbreviations 118

xi ABSTRACT

Time of origin has been determined for neurons of the anterior

olfactory nucleus, olfactory tubercle, nucleus of the lateral

olfactory tract, islands of Calleja, and preoptic and septal nuclei.

Pregnant mice were injected intraperitoneally and postnatal mice

subcutaneously with H 3 -thymidine once during development. Animals were sacrificed as adults; brain sections were coated with NTB-2

emulsion, exposed, developed, and stained routinely. Autoradiograms

showing the exact location of heavily and lightly labelled cell nuclei were prepared and the number of heavily labelled neurons in each nuclear area plotted against the day of development.

Early studies of proliferation and differentiation in the neural tube are reviewed, emphasizing nomenclature and also those controversies resolved with modern techniques, especially autoradiography. These matters include ventricular cells as a single cell type whose nuclei undergo to-and-fro migration during the mitotic cycle; the rounding up of ventricular cells prior to division; the ventricular zone as site of origin of most neuroblasts; subventricular cells as precursors of specialized neurons (granule cells); and some details of migration.

Conclusions and interpretations of regional autoradiographic studies, especially in the mouse, are presented with attention to gradients in time of origin, kinetics of mitosis, and general principles of neuron origin. The history of autoradiography, its physical basis, advantages, and limitations are also included.

xii xiii

Times of origin essentially agree with preliminary findings,

except in the anterior olfactory nucleus; its neurons originate longer

than previously reported. With few exceptions, the data strengthen the

general principles: that small neurons arise before larger ones; that

nuclear areas have individualized times of origin; and that sequence of

origin roughly parallels phylogenetic trend, as noted in the septal

area. Apparently opposite mediolateral and lateromedial gradients in

septal and preoptic areas, respectively, are both "outside-in", as

indeed a rostrocaudal septal gradient might be if viewed from the

sagittal aspect. "Inside-out" gradients characterize the periventric­

ular preoptic region and olfactory tubercle. The investigation supports

the impression of an earlier worker that neurons of nuclear areas arise

in an "outside-in" sequence, whereas those of laminar structures, with

a possible exception in area dentata, originate "inside-out". A

dorsoventral gradient in the periventricular preoptic region resembles

that reported elsewhere in hypothalamus. The full significance of these

and other gradients is not known, even though the patterns often match

gradients in differentiation and nuclear recognition, as noted in the preoptic area. The sequence and spacing of neurogenetic events varies; recent studies suggest that local factors might initiate neuron differentiation.

Although most basal forebrain nuclei have an early onset of neuron origin (from the ventricular zone), the prolonged proliferation of neurons for the pyramidal layer of the olfactory tubercle, posterior and lateral parts of the anterior olfactory nucleus, islands of Calleja, xiv and nucleus accumbens implicates a subventricular site of origin for their late-forming cells. Examples of postnatal neurogenesis, dis­ covered here and elsewhere, indicate a role in behavioral plasticity.

Time of origin of nucleus accumbens resembles that of striatal complex more than that of septal area, embryological evidence supporting other criteria which relate accumbens more closely to striatum than septum.

Times of neuron origin for preoptic nuclei were compared to those of neighboring regions, to help classify the preoptic area as telencephalic or diencephalic. These times of origin, however, resemble those in both brain chambers in one way or another, so no conclusion could be reached on this base alone. The need for a telodiencephalic boundary was examined on the basis of cytoarchi- tectonics, circuitry, and functional criteria and it was concluded that separation of the adult forebrain into telencephalon and diencephalon, except for purely topographical purposes, is unjustified. CHAPTER 1

IIWRODUCTI ON

One of the major applications of autoradiography to the study of the developing nervous system is tracing the life history of neurons from their site of origin to their final location. If injected on a specific day of development, H -thymidine is incorporated into newly synthesized DNA. of proliferating neuronal precursors. Labelling is permanent since neuroblasts never undergo subsequent division once they leave their site of origin (ventricular and subventricular zones or their derivatives). Their final position can easily be determined by examining autoradiograms of adult brain sections. Site of origin can also be established, by sacrificing animals at varying intervals after injection and following labelled neurons along their migratory paths. Such tracing of'embryonic cells has been a classical endeavor of embryology but until the application of autoradiography it was virtually impossible in the neural tube. Migrating neuroblasts often take complicated routes and are usually undifferentiated and therefore indistinguishable from glial cells and the myriad of neuroblasts migrating in different directions..

Since its first application to central nervous system histo­ genesis in 1959, autoradiography has been used to determine the time of origin (the day on which the neuronal precursor underwent final division) in many brain regions, especially of the mouse. Time and site of origin

1 2 of pontine and medullary neurons in the mouse have been studied exten­ sively (Taber 1963, Taber Pierce 1966, 1967, 1970). The cerebellum has been investigated by more people than any other area, since its circuitry is well understood and its development, from two opposing germinal layers, provides possibilities for understanding the adult anatomy (Uzman 1960a, 1960b; Miale and Sidman I96I; Sidman and Miale

1959, i960; Fujita 1967; Taber Pierce and Sweet 1967). The midbrain has received scant attention except for experimental studies in the mouse (DeLong and Sidman 1962) and histogenetic studies in the chick

(Fujita I96U) and rat (Hanaway, McConnell, and Netsky 1971). In the diencephalon, detailed studies have been reported in the thalamus and epithalamus of the mouse (Angevine 1970a) and in the hypothalamus of the rat (ifft 1972). Telencephalic areas reported in detail include the olfactory bulb (Hinds 1967, 1968a, 1968b), hippocampus (Angevine

1965), and cerebral cortex (Angevine and Sidman 1961, 1962). Other telencephalic areas of the mouse have only been reported in preliminary form (Sidman and Angevine 1962, Hinds 1967, Creps and Angevine 1971).

The present study was undertaken to establish the time of origin of many basal forebrain nuclei in the mouse, including the anterior olfactory nucleus, olfactory tubercle, islands of Calleja, nucleus of the lateral olfactory tract, and all septal and preoptic nuclei. By examining these times of origin certain developmental relationships might be established to bring order to a complex, poorly understood region of the brain. For instance, the relationship of the nucleus accumbens to either the septal area or the striatal complex is 3

•uncertain. Comparison of times of origin in these areas may show simi­

larities which could help to clarify the association of the accumbens

with septum or striatum, or at least complete an important body of

evidence upon their early history. The classification of the preoptic

area as telencephalic or diencephalic is also currently disputed.

Characteristics of telencephalic and diencephalic times and patterns

of origin should similarly be sought and compared. If the time of

origin of preoptic area resembles that of telencephalon or diencephalon,

this fact should be considered in any attempt to delineate the currently

disputed telodiencephalic boundary.

The results must also be compared to those of preliminary

studies in this brain area. The applicability of certain generaliza­ tions of time of origin recognized in earlier studies will also be

examined. These generalizations include: l) Early origin of" large

compared to small neurons; 2) Individualized times of origin in brain

nuclei as defined by cytoarchitectonic criteria; 3) Correlation of

sequences in time of origin and differentiation; 4) Correlation of sequences of origin with phylogenetic trends; 5) Postnatal origin of small neurons from subventricular cells; 6) Gradients in origin along three axes of the neural tube; and 7) "Inside-out" gradients in laminar structures compared to "outside-in" gradients in nuclear areas.

Before presenting the details of the present study, the contri­ bution which autoradiography has made to answering fundamental questions of neuroblast proliferation and differentiation will be discussed.

A brief history of autoradiography, along with its advantages and limitations, is also given. Since the complex and poorly understood areas of the basal forebrain being investigated are not commonly described in textbooks of neuroanatomy, a brief summary of forebrain tracts and nuclei in the rat (similar to the mouse) has also been included. CHAPTER 2

HISTORICAL PERSPECTIVE

Although the primary application of autoradiography to the study of central nervous system histogenesis is tracing the life history of neurons, this new tool, in combination with electron microscopy and older methods, has helped resolve controversies debated many decades by the classical histologists. They, often, had to devise a new technique before even approaching a problem; and many of their techniques are still important tools today. It is a tribute to their genius that despite their limited repertoire of instruments, they formulated problems which we have begun to solve only in the last few years. The contribution which autoradiographic studies have made is better appre­ ciated when placed in this perspective.

Proliferation in the Neural Tube

Classical Studies

The first reported study of central nervous system histogenesis was by Altmann in l88l. His observations led him to formulate the principle that in developing epithelial organs, including the neural tube, mitosis always occurs at the surface farthest from the mesoderm.

Furthermore, he observed that the mitotic spindles were always parallel to the surface. Schaper (1897) found that Altmann's principle applied

5 to a wide variety of vertebrate species. In 1936 F. C. Sauer reiterated this principle.

His (1889) made extensive studies of the developing central nervous system and identified two cell types in the early neural plate.

Most cells were in some phase of mitosis; these he called germinal cells. A few cells were not dividing. As the neural plate develops into the neural tube, His believed that the non-dividing cells, or spongioblasts, elongated, their processes uniting with adjacent processes at the luminal and mesenchymal surfaces to form the inner and outer limiting membranes, respectively.

His (1889) distinguished three concentric layers in the neural tube. Next to the lumen the germinal cells, lying between the processes of spongioblasts, comprised the columnar zone. The intermediate layer, containing the nuclei of the spongioblasts, he called the nuclear zone.

The layer adjacent to the mesenchyme, composed of the outer processes of spongioblasts, was the marginal zone. As development proceeds the germinal cells proliferate rapidly. Some of the daughter cells leave the columnar zone and migrate to a position between the nuclear zone and the marginal zone, thus forming the mantle zone. These migrating daughter cells His designated neuroblasts, an unfortunate term, since

-blast means a cell capable of division whereas a neuroblast is postmitotic. His believed that germinal cells produce neuroblasts exclusively, since he observed many forms transitional between germinal cells and neuroblasts but no forms transitional between germinal cells and spongioblasts. (For a discussion of current terminology consult

Boulder Committee 1970.)

Since the number of spongioblasts does increase during develop­

ment, even though they were considered incapable of division, Schaper

(1897) suggested that spongioblasts are also produced by germinal cells.

He believed that germinal cells and spongioblasts are simply the resting

and dividing forms of a single cell type rather than two cell types, as thought by His; this view is now known to be correct. His later agreed that Schaper was probably correct.

Schaper explained Altmann's principle by saying that mitosis occurs at the place of least resistance and nearest the source of nour­ ishment, which, -in the case of the neural tube, is the surface adjacent to the lumen. He thought that germinal cells also produce indifferent cells which migrate into the mantle zone where they differentiate into either neuroblasts or spongioblasts.

Ramon y Cajal (1909-1911) studied central nervous system histo­ genesis extensively. He agreed with Schaper (1897) that germinal cells can produce both neuroblasts and spongioblasts; but he did not agree that there are migrating undifferentiated cells„ He believed that cells differentiate "before reaching the mantle zone, since he observed transi­ tional forms in the primitive ependymal zone (columnar and nuclear zones of His) of the retina, cerebral vesicle, and spinal cord„

In 1935 F. C. Sauer presented evidence for Schaper's foresighted hypothesis (1897) that germinal cells and spongioblasts are two forms of the same cell type. He observed that the processes of neural plate 8

cells are attached to each other by terminal bars at the luminal surface

but not at the mesenchymal surface. He believed that columnar cells

withdraw their outer process, round up, and divide adjacent to the

lumen, retaining their attachments there by their terminal bars. After

division they elongate again. In support of his idea, Sauer (1936)

noted that the size and shape of nuclei adjacent to the lumen are

characteristic of cells just before or just after division, while nuclei

at deeper levels resemble those of cells during later stages of mitosis.

Sauer's important realization, that nuclei move in and out within the neural tube during the various states of mitosis, was not widely accepted for many years (Weiss 1955).

Convincing evidence for F. C. Sauer's conclusion was supplied in 1956 by Watterson, Veneziano, and Bartha. They applied colchicine to chick neural tube and showed that nuclei, blocked at metaphase, were found at progressively deeper levels the longer the block was maintained.

If Weiss and the classical histologists had been correct only those nuclei adjacent to the lumen would have been arrested in metaphase.

Further support for F. C. Sauer's concept of migration of nuclei was obtained by M. E. Sauer (his widow) and Chittenden (1959). They reasoned that since DM synthesis precedes division, increasing DM content from 2n to Un, the amount of DM could distinguish a cell preparing for division from a non-dividing cell. If F. C. Sauer were correct, nuclei with DM content greater than 2n would be found through­ out the depth of the neural tube wall. If His and Weiss were correct, however, nuclei with DM content greater than 2n would be found only 9

adjacent to the lumen. The results showed that a statistically

significant number of cells deep in the neural tube wall have a DNA.

value greater than 2n.

In addition to the primitive ependymal, mantle, and marginal

zones, a fourth zone of mitotically active cells develops between the

ependymal and mantle layers. This subventricular zone (Boulder Commit­

tee 1970) is probably present throughout the embryonic nervous system

at some stage of development, but is especially prominent in the

cerebral hemispheres where it was originally described by Allen (1912) in rats up to two years of age. Later it was studied in human embryos by Kershman (1938) who named it the subependymal zone. He described two types of cells, apolar neuroblasts and apolar spongioblasts, suggesting that this layer is a source of both neurons and glia. This has since been confirmed by autoradiographic studies (see below).

Autoradiographic Studies

Migration of nuclei within the primitive ependymal zone

(ventricular zone of Boulder Committee 1970) of the neural tube during phases of mitosis, as proposed by F. C. Sauer (1935? 1936) was not widely accepted until autoradiography was applied to neuroembryology.

In 1959 Sidman, Miale, and Feder provided evidence for such migrations by labelling the premitotic cells of embryonic mice with tritiated thymidine. Injected into the maternal circulation, the radioactive nucleic acid component passes through the placenta and is incorporated into any cell nucleus which is synthesizing DNA. Unincorporated

-thymidine is broken down or excreted rapidly; thus a single injection is a pulse label. Embryos killed one hour after injection contained labelled cells predominately in the outer half of the primitive

ependymal zone. Six hours after injection labelled nuclei were found

primarily in the inner half of the ependymal zone. After 2k hours most

labelled nuclei were found in the mantle zone (intermediate zone, of

Boulder Committee 1970); some, however, remained in the ependymal zone.

Forty-eight hours after administration, heavily labelled nuclei were found mostly in the mantle zone and lightly labelled nuclei, presumably those of cells which had divided a few times since injection, were found adjacent to the ventricle. Sidman and his colleagues concluded that DM synthesis takes place in the outer half of the ependymal zone; nuclei move to the ventricle to divide and the daughter cells, or at least their nuclei, move away from the ventricle. Some nuclei migrate within their cell processes into the mantle zone (see Golgi and electron microscopic studies); other remain in the ependymal zone to repeat the cycle of synthesis, migration, and division.

Similar results were obtained by M. E. Sauer and Walker (1959) for the chick neural tube and by Hicks et al. (1961) for the developing retina. The latter investigators found that although the rate of proliferation decreases as the embryo matures, the in-and-out migration of nuclei persists as long as the proliferative ventricular zone is present.

Autoradiographic and histological studies of the subependymal layer in the postnatal mouse (Smart 1961) showed that this layer, first described in the rat by Allen (1912), is present in the inferior horn 11

of the lateral ventricle to the olfactory bulb. In the adult,mitotic

cells were rarely seen outside this layer, which contained many pyknotic

cells, so Smart concluded that most cells produced during adulthood

degenerate. He suggested that this layer is an evolutionary adaptation

for perinatal production of massive numbers of neurons, especially to

the caudate, not possible in the restricted ependymal zone and for

postnatal production of neuroglia.

It is now recognized (Sidman 1970b) that the subependymal layer

differs from the ventricular zone in at least two ways: its nuclei do not undergo to-and-fro movements, and they continue to divide even after outward migration has begun. It is not certain whether the subependymal layer gives rise to stem cells which can form either glia or neurons or whether it contains separate glial and neuronal precursors. The subep­ endymal layer is present in all developing mammalian nervous systems at some period, persisting for months in the cerebral hemispheres but only for days in the spinal cord. Subventricular cells in the caudal part of the roof of the fourth ventricle give rise to external granule cells of the cerebellum, which in turn eventually become the definitive granule cells (Sidman and Miale i960). A similar group of cells migrate through the pons and medulla, forming the transient pontobulbar body which gives rise to inferior olivary and pontine neurons, as well as those of the cochlear nuclei (Taber Pierce 1966, 1967). Another transient structure derived from the subependymal layer, the corpus gangliothalamicus of the diencephalon of the human fetus, gives rise to part of the pulvinar

(Rakic and Sidman 1969). In the telencephalon the subependymal layer 12

is the source of granule cells in the olfactory bulb and in the dentate

' gyrus of the hippocampal formation (Hinds 1968b, Angevine 1965, Altman

1969a, 1969b).

The phases of the proliferation cycle were studied in the chick

neural tube by Fujita (1962), using cumulative labelling. He measured

1) the percentage of cells labelled after 45 minutes, 2) the time neces­

sary for mitotic cells to become labelled, 3) the time for ail cells of

the matrix zone (ventricular zone) to become labelled nonhomogeneously,

and If) the time for all cells to become labelled, homogeneously. All

matrix cells were labelled after ten hours, indicating that neuroblasts,

"which are postmitotic and hence do not synthesize DWA, migrate rapidly

into the mantle layer. He determined that the generation time is 16

hours in the mesencephalon of the six-day chick embryo with phases of

the cycle as follows: mitotic time (m) one hour; presynthetic growth

phase (G-^) 7.^ hours; synthetic phase (s) 5.6 hours; and postsynthetic

growth phase (G2) two hours. The cycle is shorter at earlier stages

of development and varies in different regions. The mitotic cycle of

the chick embryo was analyzed with a different technique by Langman,

Guerrant, and Freeman (1966). In the neural groove stage generation

time was eight hours, DHA-synthesis five hours, and mitosis 25 minutes.

In the mouse, Atlas and Bond (1965) determined the length of

phases of the mitotic cycle by plotting the percentage of labelled

mitotic cells against increasing lengths of time from injection to 13

sacrifice. In the 11-day mouse"*" generation time was 11 hours, with M

one hour, G1 3.5 hours, S 5-5 hours, and Gg one hour. The cell cycle

in the ten-day mouse spinal cord is 8.5 hours, shorter than that in the

11-day mouse, while synthesis is about as long as in the 11-day embryo

(Kauffman 1966). In the postnatal mouse cerebellum Fujita (1967)

determined that generation time was 19 hours, mitosis one-half hour,

G-j_ 8.5 hours, synthesis eight hours, and Q two hours. (For further

details concerning the generation cycle, see Sidman 1970b).

Fujita (1965) recognized three broad and important stages in

the development of the chick neural tube. During the first stage matrix

cells proliferate; in the second stage matrix cells give rise either to

more matrix cells or to neuroblasts which migrate into the mantle layer;

and in the third stage matrix cells lose their proliferative ability, at

which time they differentiate into glial or ependymal cells. Although

most workers accept these three main stages, there is some disagreement

regarding the timing and overlap of these events. Fujita (196U) reported that neurons form from the third to the eighth day, while

glial cells form exclusively after day eight; whereas Kallen and Valmin

(1963) found that neuron formation begins on the fourth day and prolif­ eration of glia on the seventh day, Martin and Langman (1965) reported that neurons in the chick spinal cord form as early as the second day

1. Equivalent to a ten-day mouse of other workers (Sidman, Angevine, or the present author), since Atlas and Bond designate the day of conceptus day one, rather than day zero. and that neuroglia arise prior to the eighth day. Proliferation

does not occur in three distinct phases in the cerebral cortex and

cerebellum of the mouse, where neuroglia form before neuron production

has ceased (Miale and Sidman 1961, Angevine 1965). Indeed, recent

studies, including those of Rakic (1971) upon the role of astrocytes

in guiding inward migration of cerebellar granule cells, show that in

many brain regions the neuroglia are present from the very beginning

of histogenesis. Glial cells are not always formed directly from matrix

(ventricular) cells, but can arise from a separate layer of cells, the

subventricular zone (see above).

Recent Golgi and Electron Microscopic Studies

Although migration of nuclei within the primitive ependymal zone

was established by both classical and autoradiographic methods, the

rounding up of cells prior to division, proposed by F. C. Sauer (1935),

could not be demonstrated before Golgi and electron microscopic studies

of the developing neural tube. Berry and Rogers (1965) and Morest

(1970 ) proposed that ventricular (primitive ependymal) cells divide while retaining attachment of their external process to the external limiting membrane. However, Stensaas and Stensaas (1968) concluded, on the basis of serial reconstructions of electron micrographs of cells in the 45 mm rabbit cerebral vesicle, that dividing cells are spherical and have no processes. These authors could not identify, with certainty, the interphase germinal cell among the following cell types in the matrix layer: a typical spongioblast with proximal process anchored at the ventricular lumen and distal process reaching the

external limiting layer; a freely-arborizing spongioblast attached

to the lumen, with distal process ending between the matrix layer

and external limiting layer; and a columnar epithelial cell attached

to the lumen but with a very short or no distal process. They suggest,

however, that interphase germinal cells axe either columnar epithelial

cells or freely-arborizing spongioblasts, since it is unlikely that a

process could be withdrawn from the external limiting layer and regrown

within the time between mitotic divisions.

Matrix cells of the early neural tube, although of one type,

undergo ultrastructural changes during their generation cycle (Fujita

1966). Most organelles occupy the apical (juxtaventricular) cone in

those cells whose nuclei are moving toward the lumen, but occupy the

peripheral cone in cells whose nuclei are moving away. Furthermore,

the cilium and basal body disappear during Gg, mitosis, and early ;

cilia may not reform at all upon cells in rapidly proliferating areas

such as the spinal cord.

A more recent study of ventricular (matrix) cells by Hinds and

Ruffett (1971) concerned cytological changes throughout the entire

generation cycle and employed Golgi and electron microscopic methods.

They observed that prophase and prometaphase cells possess some type of external process spanning the thickness of the brain wall, whereas the rounded metaphase cells have only short bulbous processes, if any.

Anaphase and early telophase cells never have a distal process. Late telophase and early interphase cells have a distal process, which usually does not reach the external limiting layer, with end-feet which

resemble growth cones. Hinds and Ruffett believe that ventricular cells

round up just prior to anaphase and telophase and regrow an external

process just before becoming mature interphase cells. They explain the

freely-arborizing spongioblast and columnar epithelial cell of Stensaas

and Stensaas (1968) as late telophase cells reforming their external processes. While Stensaas (1967a) stated that the terminations of freely-arborizing spongioblasts do not display the fine structure of growth cones, he conceded that they were growing processes.

Rounding up of a ventricular cell could be accomplished by withdrawal, separation, or atrophy of its external process. In the presumptive olfactory bulb and paleocortex Hinds and Ruffett observed isolated processes attached to the external limiting layer, suggesting separation from the cell body prior to division. In the neocortex, where the external process is not as long, no isolated processes were seen, suggesting withdrawal or atrophy. While Fujita (1963) proposed that migration of the nucleus toward the ventricle is secondary to the withdrawal of the external process, Hinds and Ruffett observed nuclei at the lumen before any rounding of the cell, suggesting active migration. They discuss the possible involvement of microfilaments in this nuclear migration.

Hinds and Ruffett also noted that the axis of the mitotic spindle is almost always parallel to the luminal surface, as described by Altmann in l88l and others more recently. They did see, however, occasional spindles perpendicular to the surface. The possibility that 17 perpendicular spindles might facilitate liberation of neuroblasts from the ventricular zone (Langman, Guerrant, and Freeman 1966) is rejected by Hinds and Ruffett, who consider perpendicular spindles neither necessary nor sufficiently numerous to account for neuroblast release.

Differentiation of Heurons

Classical Studies

According to His (1889), neurons come from germinal cells in the columnar (ventricular) zone. Schaper (1897), however, did not observe any differentiation of cells before they reached the mantle zone and thought that neurons arise from wandering indifferent cells capable of mitosis, outside the primitive ependymal zone. Ramon y

Cajal (1909-1911; see also i960), using reduced silver to demonstrate neurofibrillae, studied the differentiation of germinal cells into neuroblasts within the ependymal zone. He described four phases of the developing neuroblast: apolar, bipolar, unipolar, and multipolar.

The first indication of differentiation is the appearance of a neurofibrillar network in the distal cytoplasm of apolar neuroblasts, in or near the germinal zone. Apolar neuroblasts usually develop two opposing prolongations and become bipolar neuroblasts. Bipolarity is usually attained while the neuroblast is migrating from the ependymal to the mantle zone. In most cases the distal process is larger and richer in fibrils than the proximal one. As the bipolar neuroblast approaches the mantle zone the distal process, the primordial axon, becomes longer and thicker, with a thickening, the growth cone, at its 18

tip. Usually the proximal process atrophies, leaving a unipolar

neuroblast. The multipolar neuroblast forms as cytoplasmic projections

grow out from fhe cell body. Cajal (see Ramon y Cajal) noted that the

apolar stage is easily demonstrated in chick retina and anterior cerebral

vesicle, but is rarely seen in the spinal cord. Windle and Baxter (1936)

also found the apolar phase infrequent during very early stages of

neuroblast differentiation in the rat brain stem.

Ramon y Cajal concluded from his silver work and other studies

using his gold chloride method for astrocytes (1913) that neurons arise

exclusively from germinal cells within the ependymal zone. He proposed

that the extraependymal indifferent cells observed by Schaper were

groups of apolar cells which he called the "third element." Rio-Hortega

(1918) impregnated this third element with silver carbonate; it proved

to be not a wandering indifferent cell, as thought by Schaper, but to

consist of two types of neuroglia, which Rio-Hortega (1920) called the

oligodendroglia and the microglia. Even though Cajal and Rio-Hortega had demonstrated that proliferative cells outside the ependymal zone were neuroglial, not neuronal, some still believed in the presence of wandering indifferent cells (Bailey and Cushing 1926).

Kershman (1938) used various reduced silver methods in human embryos to confirm the presence of apolar neuroblasts differentiating in the primitive ependymal zone. All apolar cells outside the ependymal zone developed into astrocytes or oligodendroglia, never into neurons.

He did find, however, true undifferentiated cells, or medulloblasts, in the external granular layer of the cerebellum, consistent with the fact

that medulloblastoma tumors are limited entirely- to the cerebellum.

A specialized group of mitotic cells outside the ependymal zone,

the subependymal layer, does produce neurons and glia (see Proliferation

- Classical Studies). It is not known, however, whether the cells of

this layer are truly indifferent, i.e., whether a single cell gives rise to both neurons and glia.

Electron Microscopic Studies

Ultrastructural studies have established conclusively "the nonsyncytial, pseudostratified columnar structure of the neural tube. The interphase cell (spongioblast of His and Stensaas) has "been described in detail by Lyser (196U, 1968) and Hinds and Ruffett (1971).

An elongated nucleus with one or two nucleoli occupies the widest part of the cell. Cytoplasm is found mostly around the nucleus, in apical and basal cones which narrow to form inner and outer processes, respectively. Within the cytoplasm are polysomes, granular endoplasmic reticulum, a Golgi apparatus (in the apical cone), a pair of centrioles, and scattered mitochondria. Microtubules, oriented parallel to the long axis of the matrix (ventricular) cell and predominating in the apical process, were described by Herman and Kauffman (1966). The plasma membrane, which completely surrounds the cell, has cilia at the luminal surface. Terminal bars at the luminal surface, described earlier "by

F. C. Sauer (1935), include zonulae adhaerentes and gap junctions.

External processes do not possess junctional specializations but simply 20

oppose one another. A continuous basal lamina covers the outer surface

of the neural tube.

The ultrastructural characteristics of a differentiating

neuroblast, however, have not been agreed upon, even though classical

histologists, using silver impregnation, regarded differentiating cells

as those which contained neurofibrils (see below). Fujita and Fujita

(1963) considered the first ultrastructural indications of differ­ entiation to be the appearance of granular endoplasmic reticulum and indentation of the nucleus, changes observed in cells outside the matrix

(ventricular) zone. Lyser (196U, 1968) described cells differentiating within the ventricular zone of chick embryos 'which she identified as neuroblasts by the presence of an axon. At this time (two days of incubation) these cells retain internal and external processes and resemble the bipolar neuroblast described by Cajal (1909). The cell body contains areas devoid of organelles except for numerous filaments

(6O-I3O A wide). A day later these cells lose their processes and the cytoplasm no longer contains small filaments, which have probably streamed into the growing axon. By the next day the neuroblast is multipolar, with numerous dendrites. The cell body contains most of the adult organelles in reduced numbers; scattered thick filaments

(probably microtubules) are present but appear larger than those seen during the bipolar stage. During the remainder of development the number of organelles (particularly the amount of granular endoplasmic reticulum) increases. Further differentiation depends upon growth of the axon into mesodermal tissues, possibly indicating interactions 21

between the growing axon and the mesoderm. (For details of later

cytodifferentiation, see Lavelle and Lavelle 1970). The presence of

neurofibrils in apolar and bipolar cells has been used, as previously

mentioned, by early histologists (Cajal 19H5 Windle and Baxter 1936,

Windle and Austin 1936) to identify differentiating neuroblasts.

Sechrist (1969), using silver impregnation, has described a neurofibril­

lar network corresponding to aggregates of 60-100 2 neurofilaments seen

at the electron microscopic level. These neurofilaments were observed

not only in postmitotic apolar and bipolar cells during early stages of

development, but also in some premitotic apolar cells, indicating that

differentiation may occur prior to division. Hinds and Ruffett (1971),

however, have seen neurofilaments in cells during all phases of mitosis

and suggest that they may be related to the outgrowth of the external process of the ventricular cells rather than an indication of neuroblast differentiation. Sechrist (personal communication 1972) agrees that all

ventricular cells in the late telencephalon of the mouse (as studied by

Hinds and Ruffett) may contain neurofilaments, but noted that in earlier stages of the brain stem and retina prominent neurofilamentous networks, probably do characterize differentiation since these occur only in those cells presumed to give rise to axons, just prior to and during early formation of the mantle layer, i.e., concomitant with initial neuroblast production. Sechrist (1969) also noted that apolar cells containing neurofibrillar networks occur in pairs within the germinal zone. This supports the view (Sau.er 1935) that both daughter cells of a single division differentiate together. 22

Recent Golgi Studies

Stensaas (1967b) has described neuroblast differentiation in

the developing rabbit cerebral vesicle and shown that neuroblasts have

distinctive morphologies depending on their distance from the ventricle.

Cells nearest the ventricle have round or elongated somata, resembling

the bipolar neuroblast of Cajal (1909); axons ascend toward the cortical

plate (presumptive cerebral cortex) or turn at the surface of the

ventricular zone and run horizontally. Cells just outside the ventri­

cular zone are stellate; their axons either descend and turn horizontally-

near the matrix layer (ventricular zone) or ascend toward the cortical

plate. This sequence of changes in neuroblast morphology depending on

distance from the ventricle suggests to him that the somata rotate 180°

around an axis parallel to the luminal surface (see his fig. 9) as they

migrate to higher levels. A process which forms opposite the axon, the

praepex, always projects in the direction of neuroblast migration;

Stensaas interprets this as a process directing migration.

Morest (1968a, 1968b, 1969a, 1969b, 1970) observed that most

neuroblasts are attached to both surfaces of the neural tube when they

begin migration and differentiation. In the medial trapezoid nucleus

(Morest 1969a) the neuroblast detaches its internal process and its

nucleus migrates toward the pia. The primitive internal process usually

becomes the axon, developing a growth cone and extending toward its

synaptic site. Dendrites begin to form as the external process detaches

from the external limiting layer. Dendritic differentiation never occurs before the internal process enters the region where the afferent 23

axonal plexus is forming, but always coincides with axonal growth into

the region. Morest thinks that the ingrowing axons may induce dendritic

growth by chemical factors secreted by the axons, as suggested by Cajal

(1893), or by providing a framework for the growing processes, as

postulated by Weiss (193*0.

Morest(lhas emphasized several points about neuronal

differentiation, which although not especially new, are valuable gener­

alizations: l) Dendritic growth cones are reliable indicators of growth

and differentiation; 2) Dendrites of large neurons usually differentiate

before those of smaller ones; 3) Golgi type I neurons generally differ­

entiate before type II, peripheral neurons before central ones, and phylogenetically older neurons before newer ones, although notable exceptions occur; 4) Gradients of differentiation occur in some, but not all, populations of neurons; and 5) Dendritic and other synaptic surfaces differentiate in conjunction with the afferent endings destined to synapse with them, as noted also by Purpura (1967).

In the cerebral cortex of opossum pouch young the internal process is retracted as the nucleus migrates to the mantle zone (Morest

1970). When the nucleus reaches the external limiting layer the external process detaches and disappears. Dendrites differentiate; apical ones form before basal ones, just as they do phylogenetically.

The nucleus migrates to the site of origin of the axon, establishing the adult structure. The migration of the nucleus, Morest believes, is simply a shift of nucleus and cytoplasm within the membrane rather than a migration of the entire neuroblast. 2k

Morest (1970) also considers the matter of developmental gradients. In the cerebral cortex, dorsoventral and rostrocaudal gradients of differentiation were observed. In the hippocampus and caudate nucleus, axons and dendrites differentiate before the nucleus begins to leave the ependymal zone. Gradients of differentiation within the caudate proceed rostrally and caudally from the anterior commissure. In the olfactory bulbs, the order of differentiation of the different cell types corresponds to the order in time of origin observed by Hinds (1968a), but this may not be true for other areas of the brain. In this regard, Angevine (1970a) discusses the important distinction between cytodifferentiation and nuclear differentiation, i.e., aggregation of cell bodies into the nuclear regions recognizable in mature brains. The interval between these two events varies considerably, depending on the distances involved in migration and other factors (see Discussion). CHAPTER 3

AUTORADIOGRAPHIC STUDIES OF REGIONAL HISTOGENESIS

Spinal Cord

Fujita (196U) determined the time of origin of neurons in chick

spinal cord by continuous thymidine labelling from a given day of

incubation. Neuroblasts synthesizing DNA prior to onset of labelling

are nonradioactive, thus the time of origin for all neuroblasts could

be determined. Ventral horn neurons originate early (three to six days

of incubation). Large neurons of the medial columns arise first, those

of the dorsal horn later (six to eight days). Dorsal horn neurons come

from matrix cells of the alar plate while ventral horn neurons appar­

ently arise from alar and basal plates. Lateral horn neurons probably

arise from alar plate simultaneously with dorsal horn cells. No

gradient in the rostrocaudal axis was found. Fujita (1965) noted that

gliablasts migrate to the subpial region where they produce astrocytes

and oligodendrocytes. Recent studies by Kanemitsu (1971) examine overlap of neuronal and glial time of origin and (1972) detail regional distribution of neurons arising at different stages.

Myelencephalon

Taber began in 19&3 a systematic survey of the time of origin of brain stem neurons in the mouse. A single injection was used to pulse label only those cells synthesizing DNA at the time of injection. She

25 26

observed that many neuron antecedents undergo final division prior to

the tenth day of gestation (E10), all of them arising prior to day lU.

Most neurons destined for motor cranial nerve nuclei, inferior olivary

nucleus, and reticular formation arise on or before day ten. Secondary

sensory neurons arise on days ten and 11. Cochlear neurons, lateral

cuneate neurons, and neurons of the nucleus raphe pallidus arise on day

12. The majority of neurons of each nucleus undergo final division

within a 48 hour period. A caudorostral gradient was observed for the

nucleus gracilis (caudal neurons arise before rostral ones). Brain stem

neurons with similar connections do not necessarily arise concomitantly.

Subsequently Taber (Taber Pierce 1967) studied histogenesis of

the ventral and dorsal cochlear nuclei. Neurons destined for these

nuclei arise according to size: large neurons from E10 to E12, medium

to small neurons from Ell to ElU, and granular neurons from E12 to

postnatal day seven (P7). Postnatal origin of granular neurons of the

ventral cochlear nucleus was also recognized by Altman and Das (1966).

In her article, Taber Pierce confirmed the site of origin of

cochlear neurons from the rhombic lip of the lateral recess of the

fourth ventricle, in a modern counterpart to Essick's monumental

analysis of the fate of the rhombic lip in 1912. The rhombic lip

first appears on E12. On E13 a secondary lip, containing many labelled

cells dividing from E15 to El8, appears between the primary lip and the

roof plate of the fourth ventricle. On ElU, in part of the rhombic lip known as the acoustic area, an external granular layer appears, contin­

uous rostrally with the external granular layer of the cerebellum and caudally with the pontobulbar body. Thus the acoustic external granular layer functions as an extraventricular proliferative matrix. Histo­ logical and autoradiographic evidence suggests that the primary rhombic lip gives rise to large cochlear neurons and the secondary rhombic lip produces granular neurons.

The source of cochlear neurons lies caudal to the part of the rhombic lip which generates the external granular layer of the cere­ bellum and rostral to that which produces neurons of the inferior olivary complex, raphe nuclei, lateral reticular nucleus, lateral cuneate nucleus, and griseum pontis. Within this acoustic area the caudal part gives rise to the neurons of the dorsal cochlear nucleus and the rostral part to the neurons of the anterior ventral cochlear nucleus. Between these two parts is the site of origin of the neurons of the posterior ventral cochlear nucleus.

The proliferative cycle in the myelencephalon, according to a study in the rat medullary ventricular zone by Ellenberger, Hanaway, and

Netsky (1969), is similar to that described in other regions (12 hours).

The time of origin of the medial accessory olive is ElU-15; that of the principal nucleus and dorsal accessory nucleus ElU. The site of origin is the entire alar plate; cells migrate through mantle and marginal layers to ventral parts of the inferior olivary complex where they displace dorsally cells previously formed. A true secondary rhombic lip, as described by His (1890), is not present in the rat medulla but the term is often used to describe the lateral parts of the alar plates. Metencephalon

Pons

Taber Pierce (1966) presented an interesting comparison of the

histogenesis of four nuclei of the pons: nucleus griseum pontis, nucleus

corporis pontobulbaris, nucleus reticularis tegraenti pontis (Bechterew),

and the nucleus raphe pontis. Neurons of the nucleus griseum pontis

and nucleus corporis pontobulbaris arise over several days of gestation,

from 12 to 16, with a peak of proliferation on days 13 and 14. Neurons

for nucleus reticularis tegmenti pontis arise over a somewhat shorter

time, on Ell through Ell+, with a peak on E12 and E13. The nucleus raphe

pontis originates abruptly on E10 and Ell.

In this study Taber Pierce followed the migration of labelled

cells and again corroborated Essick's classical demonstration (1912)

that the nucleus griseum pontis and nucleus corporis pontobulbaris both

arise from the pontobulbar body. This transient structure shows limited

mitotic activity and contains neuroblasts migrating from the primitive

ependymal zone to these nuclei. The site of origin of the nucleus

reticularis tegmenti ponti and nucleus raphe pontis is a limited area

of primitive ependyma in the basal plate at the level of the pontine

flexure. In her earlier paper Taber (1963; see Myelencephalon) had already shown that an area of rhombic lip caudal to the lateral recess produces not only neurons of the inferior olivary complex and of the raphe but also a few cells which migrate to the pontobulbar body and eventually reach the nucleus griseum pontis and nucleus corporis pontobulbaris. 29

Sensory nuclei of the train stem studied in the mouse "by

Taber Pierce (1970) include those of the trigeminal nerve. In the metencephalon, the principal sensory nucleus arises from E10-13, along with the caudal subnucleus of the spinal nucleus in the medulla. The oral and interpolar subnuclei have a slightly shorter span of origin,

E10-12, and the mesencephalic nucleus arises quite early, on E9« A caudorostral gradient in time of origin was found in the principal sensory nucleus and in the spinal nucleus, subnucleus caudalis, and a mediolateral gradient in subnucleus caudalis.

Cerebellum

A plethora of autoradiographic studies of cerebellar histo­ genesis have appeared since i960. The detailed knowledge of adult anatomy makes the cerebellum an attractive area for studying development. Despite this, inconsistent conclusions have been reached by different investigators. Uzman's studies (1960a, 1960b) in the mouse led him to conclude, as others later confirmed, that Purkinje and Golgi type II cells (the oldest of the cell types) arise early, prior to day 13. He believed, however, that these cells arise from the subependymal germinal layer and, after migrating outward through the mantle layer, contribute to the external granular layer, later separating from it to take up their adult positions. The other cells of the external granular layer proliferate in situ (as evidenced by radioactive mitotic cells) and produce small nerve cells and glia which then migrate inward to the internal granular and molecular layers. 30

According to Uzman, the molecular layer also contains cells which arise within the internal granular layer.

Uzman attempted to determine the kinetics of proliferation in different parts of the cerebellum and showed that the external granular layer has a much greater proliferation rate, as evidenced by greater dilution of label, compared to the internal granular layer. He deter­ mined that nine days after birth, cells in the external granular layer divide once every two hours; cells in the middle part of the internal granular layer divide two-three times in 2k hours; while most cells in the outer part of the internal granular layer beneath the Purkinje cell layer do not divide even once in 2k hours.

In later studies of the developing mouse cerebellum, the Sidman group (Male and Sidman 1961, Sidman and Male 19593 19&0) concluded that Purkinje cells (and presumably Golgi type II cells) never occupy a position in the external granular layer, as proposed by Uzman, but arise instead from the primitive ependyma in common with the neurons destined for the deep nuclei of the cerebellum. According to them, Purkinje cells (as Uzman thought) and neurons of the deep nuclei arise on Ell-13-

Golgi type II cells, however, were said to arise later, from E12-15.

Finally, Altman and Das (1966) and Altman (1966), in their studies of histogenesis of cerebellum and other areas in postnatal rats, described occasional radioactive Golgi type II cells and concluded that these cells can arise after "birth, therefore from the external granular layer.

Miale and Sidman (1961) demonstrated that in the mouse the external granular layer forms on E13 by migration of proliferative cells 31

from the ependymal zone of the fourth ventricle over the surface of the

cerebellum. Proliferation of these cells peaks during the first

postnatal week and fades away by the end of the third week. Miale and

Sidman agreed with Uzman that the cells of the internal granular layer

divide more slowly than those of the external granular layer, but found

the generation time in the external granular layer was nearly 2h hours,

not two hours as determined by Uzman. They also concluded that all

granule, stellate, and basket cells arise from the external granular

layer, not from the internal granular layer as proposed by Uzman.

Altman and Das (1967) showed that in the guinea pig, a precocial mammal, the external granular layer is thinner than in the mouse and virtually disappears by the sixth postnatal day.

Fujita (1967) strengthened the conclusions of Miale and Sidman concerning the fate of the external granular layer. Using cumulative labelling of a ten day mouse, Fujita demonstrated that after ten hours only the outer cells of the external granular layer were labelled, and after 28 hours virtually all cells were labelled, along with a few cells in the molecular layer. After 31 hours many cells in the internal granular layer were radioactive. He concluded that it takes a cell 28 hours to proceed from DMA. synthesis in the outer half of the external granular layer to the postmitotic stage at the inner edge, while it takes only three hours for it to traverse the molecular layer and enter the internal granular layer. In another study of the proliferative kinetics of cells in the external and internal granular layers Fujita,

Shimada, and Nakamura (1966) found that the generation time varied on 32

different postnatal days, decreasing from 29 hours on postnatal day one

(Pi) to 15 hours on P7, and increasing again to 21 hours on P10.

Fujita (1967) also studied, as Uzman and Sidman's group had,

various cell types in the cerebellar cortex of the mouse, with similar

results. He found that Purkinje cells and Golgi type II cells arise

prenatally; granule cells of the vermis arise from Pl-15 and those of

the hemispheres arise from PU-15. Stellate and basket cells arise

before granule cells, originating largely before the end of the first

postnatal week.

In the chick Fujita (I96U) found that the first cells to

differentiate (three-four days of incubation) are the large cells

destined for the midline deep cerebellar nuclei. Most neurons of the

deep cerebellar nuclei have arisen by the fifth day of incubation.

Purkinje cells arise on the fifth to sixth days, those of pyramis and

middle lobe before those of nodulus and anterior lobe. Golgi type II

neurons arise from the matrix layer on the sixth and seventh days of

incubation. These findings are consistent with his findings in the

mouse.

In the mouse the time of origin of deep cerebellar nuclei was also studied by Taber Pierce and Sweet (1967). Small and medium-sized neurons have a slightly longer period of origin, beginning on E10 and ending by ElU, compared to larger neurons. Purkinje cells arise on E12 and apparently migrate past the deep cerebellar neurons, as described by Miale and Sidman (1961). Fujita attempted to determine whether the cells of the external granular layer were wandering indifferent cells, as proposed by Schaper

(1897)5 or whether they produce neurons exclusively. Uzman (1960b) had concluded that both neurons and glia were produced there, while Miale and Sidman (1961) believed that glia were produced largely in the ependymal zone. Fujita showed that the number of labelled cells in the external granular layer increases from 16 to 21 days after labelling.

This increase he considered too great to be due to neuron production alone; thus he supported the view that the external granular layer is an ex-traependymal proliferative zone equivalent to the matrix layer containing true wandering indifferent cells as proposed by Schaper.

Fujita proposed that proliferation in the external granular layer has four distinct phases, similar to the three phases in the matrix layer (see Autoradiographic Studies). During stage 1 the cells produce other cells of their own type; in stage 2 they produce stellate and basket cells; in stage 3 granule cells; and during stage k glial cells.

Altman (1969a, 1972a, 1972b, 1972c) studied postnatal cerebellar histogenesis in the rat using autoradiography and a variety of other methods. He found (1969a) that the cortex increases in volume from birth to 21 days. During the first week the external granular layer produces basket cells, stellate cells, and granule cells, which differentiate in that order. No labelled Purkinje cells were seen postnatally but, as in previous studies (see above), an occasional Golgi type II cell was seen, indicating that not all form before birth. 3^

Several regional differences in rate of development were noted; the

sulci were more highly developed than the gyri; the ventral lobes

(nodulus and lingula) precede the anterior lobes in development; and

the hemispheres generally differentiate after the vermis.

Studying the external granular layer and histogenesis of its

cell types, Altman (1972a) found that the granule cell bodies migrate

downward when the parallel fibers reach their final locations. The

result is a piling up of parallel fibers in the molecular layer, the

first formed fibers nearest the outside. He stated that basket cells

are arrested in the molecular layer because their processes are perpen­

dicular to the bed of parallel fibers. Basket cells line up according

to their sequence of differentiation within the molecular layer.

Parallel fibers were seen to synapse with basket cells even before they established synapses on spines of Purkinje cells.

Altman also looked at the sequence of synapse formation in the various cell types in the cortex. He demonstrated that after neuronal synapses are formed remaining nonsynaptic surfaces of neurons are invested with glial processes. Synapses on Purkinje somata form before dendritic synapses and synapses in the inner molecular layer form prior to those in the outer molecular layer (1972b). He then showed (1972c) that mossy fibers synapse with Golgi type II cells, and Lugaro cells with recurrent collaterals of Purkinje cells. He concludes that there are three major stages in development of cerebellar cortical cells: morphogenic, synaptogenic, and gliogenic. 35

Miale and Sidman (1961) wondered whether synaptic connections in

the cerebellar cortex might be established during migration, when two

cells are adjacent to one another, a thought expressed also by Angevine

and Sidman that same year for the cerebral cortex. For instance,

external granule cells migrate inward at the same time that Purkinje

cells are elaborating dendrites; synapses between granule cell axons

and Purkinje cell dendrites might be established at this time. Simi­

larly, neurons of the deep nuclei might be in contact with Purkinje

cells at early stages, since they have a common site of origin and

could even be daughter cells of a single mitotic division. Altman

(1972b, 1972c) found no evidence to support this interesting speculation

which remains open to experimental test, perhaps in tissue culture.

Rakic and Sidman (1970) analyzed the histogenesis of the human

cerebellum, particularly the lamina dissecans, a transient acellular

band between the internal granular layer and the layer of Purkinje

cells. Prior to ten weeks the only proliferative zone is at or near the ventricle; but afterward the external granular layer forms and

becomes the principal proliferative zone. Purkinje cells and neurons of the deep nuclei form before the migration of external granular cells, so presumably they form in the ventricular zone. The possible formation of synapses during migration discussed in an earlier paper (Miale and

Sidman 1961) "was not supported here, since Purkinje cells originate before any granule cells have formed. The sequence in time of origin is not necessarily the same as the sequence in time of differentiation; although Purkinje cells and neurons of the deep nuclei arise at the same 36 time, the former differentiate much later. Rakic and Sidman conclude that the presence of a lamina dissecans, found only in the cerebella of humans and whales, signifies the prolonged interval between early

Purkinje cell and late granule cell development. The distinctive cell-sparse band represents a layer of afferent endings waiting for formation of their target neurons.

Mesencephalon

A study by DeLong and Sidman (1962) concentrated, not upon details of regional development, but upon the effect of eye removal on histogenesis of the superior colliculus in the mouse. In normal mice, neurons destined for the superior colliculus originate in the ventri­ cular zone on the 11th day of gestation and most of the neuroglia arise during the first postnatal week. Removal of one eye at birth causes some neuronal degeneration and decreased glial production during the first week. A gliosis develops after two weeks.

Fujita (196^-) studied differentiation of the optic tectum of the chick using cumulative labelling and found that the inner layers, stratum griseum centrale and stratum griseum periventricularis, origi­ nate first, from five to six days of incubation, the dorsomedial part somewhat earlier than the ventrolateral. The stratum griseum super- ficiale originates later, from the sixth to the ninth day of incubation.

The cells apparently come from the ventricular zone and migrate past the earlier formed cells to reach their superficial location. Within the superficial layer, however, the most superficial neurons are oldest and the ventromedial part older than the dorsolateral part. Time of origin of the substantia nigra, ventral tegmental area

of Tsai, and interpeduncular nucleus in the rat was studied by Hanaway,

McConnell, and Netsky (1971). Large neurons arise from Ell to E15,

while granule cells and glia, although beginning early, continue to

originate much longer, nearly until birth (Ell-22). Nigral neurons

arise from the basal third of the ventricular zone and migrate radially

between previously formed cells, in an "inside-out" pattern. The

ventral tegmental area of Tsai and interpenduncular nucleus arise from

the middle third of the basal plate and migrate ventrally. These nuclei

are recognizable by the l8th day of gestation.

The time and site of origin of the subcommissural organ and

adjacent ependyma of the mouse were determined by Rakic and Sidman

(1968). Subcommissural cells arise in the ventricular zone of the rostral midbrain from Ell-15. Ependymal cells arise over a longer period, from Ell-18.

Diencephalon

In studies of mouse retina histogenesis Sidman (1961) found labelled mitotic cells from Ell to P10 at the outer surface of the retina, which is analogous to the ventricular surface of the brain.

To-and-fro movement of the nuclei occur as in the ventricular zone of the developing brain, but a few late forming cells arise locally.

Cells of all three retinal layers form simultaneously, but the relative number of cells for each layer varies on different days. The ganglion cell layer attains its prospective neuronal complement earlier than the other layers. Large ganglion cells and amacrine cells form throughout the third week of gestation (Ell-18) and small ganglion cells during the

last few days (E15-18). Horizontal cells originate on E13, "but bipolar

cells arise primarily after birth. An interesting synchrony of cell

origin was noted in the photoreceptive and bipolar layers: neurons of

the inner halves of these two layers arise on E17, while cells of the

outer halves arise in the first postnatal week. Such simultaneous

origin of certain nerve cells, known to be closely related functionally

at maturity, may mean that their synaptic connections are established

early, when the cell bodies are contiguous, as speculated in studies of

cerebellar and cerebral cortices (Miale and Sidman 1961, Angevine and

Sidman 1961). Sidman also noted that the posterior pole of the eye forms before the ora serrata and that gliogenesis occurs during the first postnatal week.

Other components of the diencephalon have received scant atten­ tion until recently. Altman and Das (1966, and ALtman 1966) observed that no neurons of the lateral geniculate form postnatally, but glia form until the 13th postnatal day, with a peak during the first week.

A remarkable translocation of embryonic components, in this case the specialized derivatives of the subventricular layer, from one brain chamber to another has been described by Rakic and Sidman (1969).

The human pulvinar continues to receive neurons, after the ventricular zone ceases neuron production, from cells of the ganglionic eminence, a huge structure in the fetal telencephalon. Cells migrate from the fifth through the eighth month from the eminence to the prospective pulvinar, passing en route through a previously undescribed transient structure, the corpus gangliothalamicus. The pulvinar is absent or

poorly developed in nonprimates, hence this migration cannot be studied

in experimental animals. but similar long-distance translocations in

other areas may be anticipated.

The time of final division of neurons of all diencephalic

nuclei dorsal to the hypothalamic sulcus was analyzed by Angevine

(1970a). The ventral thalamus arises during an early, brief period

(E10-13), paralleling its early differentiation. The dorsal thalamus

arises over a longer period (E10-15) and the epithalamus lags slightly

behind (E12-16). The sequence, ventral thalamus, dorsal thalamus, and

epithalamus, demonstrates a ventrodorsal gradient in time of prolifer­

ation. In addition, the oldest cells of the dorsal thalamus are at

caudal levels while the youngest cells lie at the most rostral level.

These and other findings indicate a caudorostral gradient, similar to

a subsequent gradient in differentiation. The habenula shows an

"outside-in" gradient, in that neurons of the lateral nucleus arise

before those of the medial (Ell-13 vs. E12-16). The final cells

produced remain just beneath the ependyma. Similarly, lateral neurons

of dorsal thalamus precede medial ones. This pattern characterizes many

other nuclear areas of the brain, whereas an "inside-out" pattern characterizes isocortex and allocortex (Angevine and Sidman 1961, 19&2;

Berry and Eayrs 1963; Berry, Rogers, and Eayrs 19&3, 196k; and Angevine

1963, 196*+5 1965) where the basic organization is columnar. In many instances, large neurons arise before smaller ones, although exceptions occur. Such a sequence, as Ramon y Cajal and others have thought, might Uo

provide the "basic framework of the nervous system and then the packing

of small cells around large ones.

Angevine also comments on time of origin in relation to other

events in neurogenesis: cytodifferentiation, migration, and nuclear

aggregation. He concludes that "...time of origin of neuroblasts

accords well with subsequent recognition of nuclei. The interval may

be only a day (geniculate bodies) or up to six days (lateral nucleus);

most nuclei, however, appear within two or three days..." (p. 1^7). He

also considers possible relationships of the notochord and neuromery to

the caudorostral gradient in the thalamus and rostroeaudal gradient in

the chick optic tectum (Cowan, Martin, and Wenger 1968).

In a recent abstract, Angevine (1970c) reports that in the mouse

many rostral hypothalamic neurons arise early (ElO-ll), middle ones later (E12-13), and caudal ones last (E15-16). This rostroeaudal gradi­

ent is opposite that seen in dorsal thalamus (1970a, 1970b). In addition, a dorsoventral gradient, also opposite that in dorsal thalamus, was noted. As in the thalamus, however, lateral neurons precede medial 1 ones.

Detailed studies of hypothalamic time of origin in the mouse have not been published (Angevine, personal communication), but a complete catalogue of preoptic and hypothalamic nuclei has been provided in the rat (ifft 1972). The pattern of origin is similar to that in the mouse although the times are slightly later (gestation is 21 days compared to 18). As in the mouse, lateromedial and dorsoventral hi

gradients were apparent. There were, however, no differences in rostral

vs. caudal levels.

Telencephalon

Olfactory Centers

Extensive studies in the mouse olfactory formation have shown

that, as in other areas, large cells are produced before smaller ones

(Hinds 1967, 1968a, 1968b). The relatively few mitral cells arise

first (EII-I8), the more numerous tufted cells somewhat later (E13-18),

and the innumerable granule cells last (E18-P20). Regional differences

in time of origin were also noted. Smaller and peripheral tufted cells

arise later than middle and internal ones, and internal granule cells

originate later and over a longer period of time than periglomerular

cells or those of the mitral cell layer. Neurons arise earlier in the

phylogenetically older accessory olfactory formation compared to their

counterparts in the olfactory bulb (mitral cells E10-E12, granule cells

E12-E18).

Neurons of the olfactory formation come from the ventricular

and subventricular zones surrounding the olfactory ventricle. The

ventricular zone, present until E17, produces mitral and some tufted

and granule cells, whereas the subventricular zone, present after ElU,

produces the remaining tufted and most of the granule cells. An

"inside-out" pattern of origin occurs; the nuclei of tufted cells migrate past those of previously formed mitral cells and, similarly, the nuclei of external granule cells migrate past those of mitral and b2

tufted cells to reach their peripheral position. Neuroglial cells

arise from E17 to P10, originating locally from proliferative precursors

derived from the ventricular zone and the layer of olfactory nerve

fibers contributed from the olfactory placode.

In a study of the extraependymal layer (subventricular zone) in

the rat, Altman (1969b) described labelled cells (tentatively identified

as granule cells and neuroglia) migrating to the internal granular

layer of the olfactory bulb. He had previously (1967) suggested that

postnatal production of microneurons might account for "neural"

plasticity or represent a substrate for "memory". Since, however, only

phylogenetically old structures exhibit postnatal proliferation, the

role of microneurons now seems to him more related "...to 'affective,

need-catering functions' ...but not with... 'cognitive instrumental

functions.'" (p. U50).

Hippocampal Region

An extensive study has been made on time of origin of the

hippocampal and retrohippocampal formations of the mouse by Angevine

(1965). Neuron formation for the region begins on E10, although only

a few neurons arise on that day. Proliferation increases rapidly on

Ell, however, and runs through highly individualized time courses for

the various component areas, in accord with the rigorous timing of

neuroblast migrations from different regions of the neural wall

(Levi-Montalcini 1963, 196*0 and the widely accepted areal boundaries

in the hippocampal region. Proliferation ceases on El8 in hippocampal

sections CA1 and CA3; sector CA2, however, attains its neuronal ^3

complement earlier, by E15, the time when neuron origin for the

subiculum and retrohippocampal formation comes to an end. In all these

areas the oldest cells are the prospective deep elements; thereafter

neurons in all components arise in a general, but not rigid "inside-out"

sequence similar to that in the cerebral cortex (Angevine and Sidman

1961, 1962), demonstrating active migration of neuroblasts (or, in light

of recent studies, their cell nuclei).

The area dentata arises also from E10-15, except for its

granular layer, neuronal precursors of which continue to proliferate

until at least postnatal day 20. The pattern of origin in this layer

is "outside-in", opposite that seen in the rest of the hippocampal region. The first clear example of a gradient in time of origin was noted in this part of the brain, beginning in the suprapyramidal limb of the granular layer and moving through the infrapyramidal limb.

Prenatally, granule cells arise from the ventricular zone; perinatally, from germinal cells near the fimbria, the daughter cells of which then migrate through CA3; and postnatally, from the granular layer itself, t where daughter cells differentiate at their site of origin. Such a matrix of proliferating cells at their definitive location had not been recognized previously.

The many regional differences in patterns of radioactive neurons noted in this study coincide remarkably well with established cytoarchitectonic divisions of the hippocampal region. For example, the prospective polymorphic neurons of the hilus of the dentate gyrus stop arising on E15, in contrast with the pyramidal cells of the CA3 kk

sector, which continue to originate through El8. Similarly, as noted

above, CA2 has a period of origin different from CA1 and CA3 which

border it. Interestingly, the latter two sectors have a fiber

connection not shared by CA2. Ontogenetic parallels of phylogenetic

trends were also considered. The postnatal origin of granule cells in

area dentata may be a developmental expression of the increased size

of stratum granulosum in placental mammals compared to nonplacental

mammals. The nearly simultaneous origin of hippocampal, pyriform,

and neocortical regions of the pallium, on the other hand, accords with the view that they are equally ancient.

In the rat, Altman and Das (1965a, 1965b, 1966) have also described postnatal formation of granule cells of the dentate gyrus.

Labelled cells were found in the basal regions of the granular layer, mostly on postnatal day ten, but some even into adulthood. Dark- nucleated, undifferentiated cells, presumed to be neuronal precursors, were sometimes in mitosis and decreased in number throughout postnatal development. Their site of origin is probably in the ependymal and subependymal layers of a prominent cell cluster, adjacent to the hippocampus, sometimes referred to as the deep root of the alveus, the elements of which incorporate label up to one month after in birth.

Similar findings in the guinea pig (Altman and Das 1967) demonstrate the postnatal origin of granule cells of the dentate gyrus of this precocial mammal. Cerebral Cortex

Studies on the histogenesis of the convexal cerebral cortex of

the mouse (Angevine and Sidman 1961, 1962) accord with the findings of

classical neuroembryology, in that cortical neurons differentiate from

the cortical plate, a peripheral lamina of cells derived from the

ventricular zone. Different cortical regions, however, are now seen

with the modern approach of autoradiography to have different times of

origin: The pyriform and entorhinal areas originate earliest (Ell-15),

whereas the cortex dorsal to the rhinal fissure arises over a somewhat

longer time (Ell-17). A caudorostral gradient in time of origin was

noted in the dorsal convex cortex, although this gradient was not nearly

as obvious as that seen in the area dentata just described (Angevine,

personal communication). Other regional differences were observed:

areas 6 and 7 differ from primary motor and sensory areas (U, 3? and 17) in showing prominent foci of supragranular neurons arising concomitantly with neurons of deeper layers; area 4 has a characteristic labelling pattern on E13 when simultaneous formation of layers iii, iv, and vi produces two prominent laminae of labelled cells. All areas of cortex studied developed in an "inside-out" sequence (see Hippocampal Region); superficial neurons arise last and migrate past earlier formed cells.

The investigators suggested that this "inside-out" pattern may facili­ tate synaptic connections which may form as neurons migrate past each other.

A similar "inside-out" pattern of neuron formation in the rat cerebral cortex was shown using X-irradiation (Berry and Eayrs 1963) and b6

autoradiography (Berry and Rogers 1965). Neurons of the innermost

layers arise on E16-17, middle layers on El8, and outermost layers on

E19-21. Berry and Rogers stressed the late arrival of the outermost

layers from both ontogenetic and phylogenetic standpoints. In their

studies of postnatal neurogenesis in the rat, Altman and Das (1966) and

Altman (1966) saw a few labelled neurons in the cerebral cortex, but

concluded that neurogenesis is primarily prenatal in this animal.

Other Telencephalic Centers

Studying only postnatal development in the rat, Das and Altman

(1970) determined that most of the neurons of the anterior poles of the

caudate nucleus and nucleus accumbens septi arise after birth (through

P5 and P6 respectively). In the mouse, striatal neurons arise primarily

before birth (E12-15; Sidman and Angevine 1962). The latter investi­

gators also reported preliminary findings upon time of origin of several

other telencaphalic nuclei in the mouse. The medial, central, and

lateral amygdaloid nuclei arise early (E10-12) while the basal and cortical nuclei lag one day behind (Ell-13). The globus pallidus also

originates early, whereas the medial and triangular septal nuclei come

later (Ell-15) and the lateral septal nucleus continues to arise nearly to the end of gestation (E13-17). In addition to confirming the fore­ going times of origin, Hinds (1967) surveyed several other areas: nucleus accumbens septi (E12-18), nucleus fimbrialis septi (E10-lU), anterior olfactory nucleus (EII-I8), olfactory tubercle (Ell-15), islands of Calleja (E13-18), nucleus of diagonal band of Broca (E10-l4), interstitial nucleus of stria terminalis (E10-15), nucleus of the b7

lateral olfactory tract (Ell-13), and massa intercalata (E12-16). The

times of origin of pyriform and entorhinal cortices were also studied "by

Hinds and Angevine (1965), who determined that the claustrum arises over

a short period (E10-13), the dorsal claustrum reaching its peak a day

earlier than the ventral claustrum. The dorsal claustrum arises concom­

itantly with the overlying basal layer of insular cortex and the ventral claustrum pari passu with the subjacent area pyriformis. Although these

and other correlations were made, the authors were unable to shed new light upon the old question of a cortical or nuclear identity for claustrum. CHAPTER k

THE BASAL FOREBRAIN

The basal telencephalic nuclei included in this study are part

of the rhinencephalon, or nose brain, since they receive primary or

secondary olfactory input. According to Papez's early view (1929) of

the rhinencephalon the olfactory mucosa, olfactory bulb, and olfactory

tubercle relay olfactory stimuli to the pyriform cortex and hippocampal

gyrus, which he considered related to olfactory sensation. The mammil-

lary body, nucleus of the stria terminalis, nucleus accumbens septi, and

parts of the caudate, putamen and amygdala integrate motor responses to

olfactory stimulation, e.g., food getting, mating, etc. Later, however,

Papez (1937) rejected his earlier view and stated that, with the

exception of the olfactory mucosa and bulb, the rhinencephalon has

little to do with olfaction, a reservation also expressed by Cajal on

various occasions. Papez proposed, instead, that the septal area,

cingulate gyrus, pyriform cortex, hippocampus, mammillary body, anterior

nucleus of the thalamus and the interconnections of these structures

provide an anatomical substrate for emotion. Broca (1878), attempting to avoid functional terms, popularized the name limbic lobe ("le grand lobe limbique") for the olfactory bulb, tract, and gyrus fornicatus on the medial wall of the cerebral hemisphere, recognizing that these

structures form a limbus or border around the diencephalon, an observa­ tion first made by Thomas Willis in l66^+ (White 1965).

^8 Sophisticated stimulation and lesioning techniques affirm the

role of many rhinencephalic nuclei in emotional behavior, as Papez

suggested. That these structures (amygdala, hippocampus, septal area,

cingulate gyrus, and hypothalamus) are also related to olfaction is

attested by the strong influence of olfaction on emotional reactions.

For farther discussion of limbic rhinencephalic terminology and history

consult White (1965), or Yakovlev (1972).

The major fiber tracts of the basal forebrain are the olfactory

tracts, anterior commissure, medial forebrain bundle, fornix, stria

terminalis, diagonal band of Broca, and stria medullaris thalami. A

description of each of these, based primarily upon the anatomy of the

rat brain, follows. A description of each basal forebrain nucleus will

be given with the results (Chapter 7).

Olfactory Tracts and Anterior Commissure

There is considerable disagreement regarding the origin of

olfactory fibers crossing in the anterior commissure. Some thought that

tufted cells of the olfactory bulbs send their axons into the anterior

commissure (Cajal 1911, Gurdjian 1925? and Allison 1953), while others

stated that no fibers from the olfactory bulbs enter the commissure

(Young 1936, Brodal 19^8, and White 1965). Recent Golgi studies in the

rat by Valverde (1965) demonstrate that axons of tufted cells do not

leave the olfactory bulbs and, therefore, could not enter the anterior

commissure; however, Ban and Zyo (1962), using the Marchi method, showed

that fibers arising in the rat olfactory bulb do pass to the contralateral anterior olfactory nucleus via the anterior commissure.

Thus, the question is currently unresolved. The origin of other major

components of the anterior commissure in the rat are less controversial;

they include: olfactory tubercle, pyriform cortex, bed nucleus of the

stria terminalis, nucleus of the lateral olfactory tract, and parts of

the amygdala (Brodal 19^-8, Valverde 1965).

There is apparently much variation as to which olfactory stria

(or tract) crosses in the anterior commissure. According to some reports it is the intermediate stria (Young 1936, Fox 19^0, Craigie, as revised by Zeman and Innes 1963); while others name the medial stria

(Kappers, Huber, and Crosby 1936). It is also often stated that the medial olfactory striae end in the olfactory tubercle, anterior hippo- campal rudiment, and septal nuclei (Gurdjian 1925). Nauta and Haymaker

(1969), Knook (1965), and White (1965),however, state that no primary olfactory fibers terminate on septal neurons.

It is generally agreed that the anterior olfactory nucleus sends fibers to the homolateral and contralateral olfactory bulbs, to the contralateral anterior olfactory nucleus, and into the homolateral lateral olfactory tract which terminates principally in pyriform cortex, amygdala, nucleus of lateral olfactory tract, and olfactory tubercle

(White 1965). Other fibers connect the anterior olfactory nucleus with frontal cortex and with the anterior hippocampal rudiment, with which it is continuous (Fox 19*+0). 51

Medial Forebrain Bundle

The medial forebrain bundle (fasciculus medialis telencephali)

is an important bidirectional link between limbic forebrain structures

and limbic midbrain nuclei, serving olfactovisceral and olfactosomatic

functions. It also directs limbic system output into the neuroendocrine

areas of the hypothalamus (Sawyer 19&7) its direct stimulation is a

positive reinforcer (Olds and Milner 195*0.

Anteriorly the fasciculus is formed by the convergence of a

dorsal and ventral bundle of fibers. The former arises in the hippo­

campus (via. fornix), septal area, nucleus of the diagonal band, frontal

cortex, and nucleus accumbens septi; the latter in the anterior olfac­ tory nucleus (pars posterior), olfactory tubercle, amygdala, pyriform cortex and nucleus of the lateral olfactory tract (Millhouse 1969).

These fibers pass through the lateral preoptic-hypothalamic region, not as a discrete bundle as the name implies, but dispersed among its interstitial neurons, the lateral preoptic and hypothalamic nuclei. All along its course numerous collaterals leave and enter the bundle. The mammillary bodies receive fibers from the bundle but do not contribute to it. Long collaterals enter the inferior thalamic peduncle and stria medullaris thalami. The dendritic trees of the lateral hypothalamic neurons pass perpendicular to the fibers and the axons bifurcate and ascend or descend along the bundle. Interstitial neurons receive information from specific telencephalic nuclei (Millhouse 1969). 52

Fornix

The fornix is the principal output of the hippocampus, but the

septal area also contributes fibers, some of 'which pass back to the

hippocampus. In fact, the septal area has been termed "...a subcortical

way station in the path of the fornix fibers. (Fox 19^0, p. 31).

Technically, the major trajectory of the fornix is intraseptal, inclu­

ding the septum pellucidum; thus, it is not surprising to find that all

septal nuclei, including the nucleus of the diagonal band of Broca,

receive fornix fibers. The precommissural fornix also passes to the

nucleus of the anterior commissure, dorsal hypothalamic area, and to

preoptic nuclei via the medial forebrain bundle. The medial cortico-

hypothalamic tract, a component of the fornix system which arises in

the caudal third of the hippocampus, projects to the periventricular

hypothalamic nuclei, including the arcuate nucleus. Dorsal fornix

fibers pass to the anterior, median, and paramedian nuclei of the

thalamus (Wauta 1956). Postcommissural fornix fibers terminate in the

mammillary complex, principally the medial nucleus. Septal fibers which

join the fornix pass to the habenula via the stria medullaris thalami,

but, in the rat at least, no fibers from the hippocampus take this route

(Nauta 1956). Only septal fibers reach the midbrain tegmentum, and no

fornix fibers could be traced to the pyriform cortex, amygdala, or

olfactory tubercle in the rat (Wauta 1956). A subdivision of the septal

area into a medial zone which is afferent to the hippocampus and a

lateral zone which is efferent, based upon study of the alligator brain

(Crosby 1917), does not apply to the rat. Stria Terminalis

The stria terminalis arises primarily in the corticomedial group

of amygdaloid nuclei, including the nucleus of the lateral olfactory

tract. It takes a circuitous route along the wall of the lateral

ventricle to its termination in the basal forebrain. A supracommissural

group of fibers pass rostral to the anterior commissure and terminates

in nucleus accumbens, anterior olfactory nucleus, accessory olfactory

bulb, and preoptic and hypothalamic nuclei (Young 1936, Gurdjian 1925,

Olmos and Ingram 1972). Recently it has been shown that many of the

precommissural fibers terminate around the ventromedial hypothalamic

nucleus (Fink and Heimer 1967). A postcommissural (preoptic) component

passes to the septal area, including the nucleus of the anterior

commissure, and also to the medial preoptic area, many fibers synapsing in the nucleus of the stria terminalis en route. Fibers arising in the

nucleus also pass to the periventricular hypothalamus via the medial

corticohypothalamic tract, also called the amygdalohypothalamic tract

(Gurdjian 1927). Another component of the stria enters the anterior commissure and terminates principally in the contralateral amygdala, but also in the nucleus of the anterior commissure, olfactory tubercle, nucleus of the stria terminalis, and pyriform cortex (Olmos and Ingram

1972). Still another component joins the stria medullaris thalami and passes to the medial habenular nucleus (see below).

Diagonal Band of Broca

This bundle of fibers, first described by Broca (l879)j forms a prominent bulge on the ventral surface of the brain passing diagonally across the anterior perforated substance. Like the medial forebrain

bundle it is almost entirely intranuclear, lying within the nucleus

of the band. Although accounts of the basal forebrain in the rat

(Craigie, as reported by Zeman and Innes 1963, Gurdjian 1925) only

mention it in passing, it has been described in detail in other animals

(Johnston 1923? Loo 19^-1» Young 1936, Humphrey 1936, and Fox 19^-0). It

is primarily a reciprocal connection between the medial septal nucleus

and the amygdala, but it also carries fibers between many other areas.

The precommissural fasciculus of Elliot Smith (1896, also known as

Zuckerkandl's bundle, 1888) descends towards and intermingles with the

ascending limb of the band and carries reciprocal fibers between the

nucleus of the band and the hippocampus, specifically fields CA3 and

CA^- (Raisman, Cowan, and Powell 1966). The horizontal limb passes not only to the amygdala, but also to the nucleus of the lateral olfactory tract, pyriform cortex, globus pallidus, nucleus of the ansa lenticu- laris, and all levels of the hypothalamus (via the medial forebrain bundle). Some axons from the nucleus of the band pass to the olfactory bulbs to modulate olfactory input (Price and Powell 1970a).

Stria Medullaris Thalami

This complex fiber bundle is the only afferent connection of the habenular nuclei. Formed at the anterior border of the diencephalon by convergence of component fibers from basal forebrain nuclei, it passes along the dorsomedial surface of the thalamus, where it has also been named the linea alba and taenia thalami (Gurdjian 1925). In the rat, which has a small habenular commissure, its principal termination is 55 the homolateral l abenula. The various components of the stria medul­ laris thalami in the rat have been described in detail by Gurdjian. A stria terminalis component carrying amygdaloid input, is supplemented by fibers from the interstitial nuclei of the stria terminalis and anterior commissure. The lateral corticohabenular tract arises in the pyriform cortex and nucleus of the lateral olfactory tract. Two olfactohabenular tracts convey information from autonomic and olfactory centers, the medial from the preopticohypothalamic area, the lateral from the olfactory tubercle, and including fibers from the medial fore­ brain bundle. Another component, the septohabenular tract, carries fibers which arise in the septal nuclei. Gurdjian (1925) and others have described a medial corticohabenular tract, fornix fibers which appeared to arise in the hippocampus, but recent studies (Nauta 1956) have failed to substantiate this. At least in the rat, all fornix fibers to the stria apparently arise in the septal area. In addition to the above fibers passing to the habenulae, others have been reported passing in the opposite direction, toward the preopticohypothalamic area. Most of these arise in the midbrain tegmentum and pass through the habenulae without synapse; whethe.r any arise in the habenular nuclei is uncertain (Nauta and Haymaker 1969). CHAPTER 5

AUTORADIOGRAPHY

Autoradiography is the technique of producing an image on a

photographic plate or film by placing it in contact with a radioactive

substance. The process is similar to photography; silver halide crystals in the emulsion are altered during exposure, thus producing a latent image. When the emulsion is developed the altered silver crystals are reduced to silver atoms, which turn Hack, thus producing the visible image. Fixer is then used to dissolve out the unreduced silver halide ions, leaving only those areas which were exposed.

Autoradiography differs from photography, however, in that the emulsion is exposed by charged particles or electromagnetic radiation emitted from a radioactive substance instead of by photons of light-

History

The first autoradiograph reported was "by Uiepce de St, Victor in 1867, who placed uranium salts on a photographic plate and thereby blackened the silver halide crystals in the emulsion. The blackening occurred even when a plate of glass was placed between the emulsion and the uranium salts. He believed that the phenomenon was one of luminescence, since the principles of radioactive decay were unknown at that time. Similarly, in 1896 Henri Becquerel placed uranyl sulfate crystals on a black paper over a photographic plate and attributed the resultant blackening to fluorescence. Two years later, the Curies,

56 interested in the foregoing studies, demonstrated the phenomenon of

radioactivity. Thus autoradiography preceded and contributed to the

discovery of radioactivity. Rogers (1967) and Hinds (1967) present

extensive accounts of the early history of autoradiography.

The first biological application of autoradiography was by

London, who in I90U produced the outline of a frog on a photographic

plate by placing the animal on the plate after exposing it to radium

vapors. The first autoradiography on human subjects was by Kotzareff

in 1922. He injected a metastatic tumor in the iliac fossa with radium

and reported that the radium slowed the course of the disease, an

observation supported recently by Greenberg et al. (1966) who showed

that tritiated thymidine is a cytocidal agent in human leukemia.

Lacassagne and Lattes (I92U) studied the distribution of polonium in

histological sections by placing a paraffin block with the radioactive

tissue against a photographic plate and then comparing the image with

the actual histological section. They coined the term autoradiography

for this process.

Actually, little biological application was made of

autoradiography before World War II because of the rarity of naturally occurring radioactive substances of biological significance and the inferior quality of emulsions for photography. After atomic reactors

and artificial isotopes were developed, biologists became interested in autoradiography as a routine technique. In 19^-0 > Hamilton, Soley, and Eichorn demonstrated uptake of radioiodine by the thyroid gland; soon after, Belanger and Leblond (19^6) demonstrated radioiodine in the colloid of the follicle and introduced the technique of applying

the emulsion directly to the slide. They scraped the emulsion from

photographic plates, melted it, and then painted it on the histological

slide with a fine brush. This method greatly improved localization of

radioactive substances by reducing the gap between the section and the

emulsion.

In 19^7 Pelc introduced the stripping film technique in which a

special emulsion is floated on water. The slide containing the labelled

section is placed under the emulsion and the emulsion and slide removed

from the water and dried rapidly. At the same time Evans (19^7)

developed the direct mounting technique in which sections are applied

directly to the photographic plate.

The next major development in the history of autoradiography was the introduction of tritium (H ) as a radioactive isotope by Fitzgerald

et al. (1951). Resolution is increased when the emissions travel shorter distances. The shorter tissue course (approximately one micron) of beta particles emitted by tritium facilitates intracellular locali­ zation of tritiated compounds. In addition, the ubiquitous presence of hydrogen in organic compounds permits labelling of all biological compounds with tritium.

The most popular autoradiographic technique, the dip-coating method, was introduced in 1955 "by Joftes and Warren. Slides are simply dipped into melted emulsion and dried in air. Several emulsions developed for this method make it a simple technique even for beginners.

The development of the electron microscope as a biological tool made 59 the use of electron microscopic autoradiography inevitable. In 1956

Liquier-Milward produced the first electron microscopic autoradiograph.

Since then considerabl. .c :.rnprovements in technique and emulsion have made electron microscopic au oradiography (Caro 1962, Salpeter 1966, and

Moses 1964) a major and popular technique. See Leblond and Warren

(1965), Rogers (1967), Zimmerman and Fautree (196B) and Baserga and

Malamud (1969) for additional details.

The Radioisotope

Every element with an atomic number greater than 83 or a mass greater than 209 spontaneously emits radiation and hence is radioactive.

The nuclei of these elements contain in a very small space many charged and uncharged particles, which are in constant motion and exert electrical, magnetic, gravitational, and intranuclear forces on each other. Such a configuration is unstable, so the nucleus breaks down

(Quimby and Feitelberg 1963). As the nucleus disintegrates it emits charged particles (alpha and/or beta) and energy (gamma waves). The rest of the nucleus rearranges itself and becomes stable.

Almost all radioactive substances emit only one atomic particle, with or without electromagnetic (gamma) waves. An alpha particle has a mass four times that of hydrogen and a charge of +2. A beta particle has a mass 1/1800 that of a hydrogen atom and a charge of -1. The rate of decay of radioactive atoms depends on the proportion of undecayed atoms present; thus the rate of disintegration is exponential. Each element has a characteristic decay rate which is expressed as a 6o

halflife, i.e., the length of time during which half of the atoms will

disintegrate.

The unit of measure of radioactivity is the curie, i.e., a

quantity of radioactive material in which the nuclei disintegrate at

a rate of 3.7 x 10^ per second. The specific activity of an isotope represents the number of radioisotopes in a mixture of radioactive and

stable isotopes of that element.

Radioisotopes not only occur naturally, but also can be induced.

In 193^ the Curie-Joliets bombarded aluminum with alpha particles; the extra mass and charge introduced into the nuclei produced the first artificial radioisotopes (Quimby and Feitelberg 1963).

The Autoradiographic Process

The basic reaction between the emulsion and radioisotope is the alteration of silver halide crystals by nuclear particles or electromagnetic radiations emanating from the radioisotope. The emulsion contains imperfections in its halide crystals called sensi­ tivity specks. When energy from ionizing radiation strikes the emulsion electrons migrate into these.specks and are trapped there. The build-up of negative charges in the crystal attracts positively charged silver ions in the vicinity; they combine to form silver atoms, releasing bromine from the crystal. When silver is formed it produces a latent image which, when developed, becomes the visible image. During development the silver catalyzes first the reduction of the entire crystal and then silver bromide in its immediate vicinity. After development time sufficient to reduce only the silver bromide near the 6i latent image, a fixer is applied which dissolves out the remaining unreduced silver bromide. The end product, a thin layer of gelatin containing silver grains, corresponds to the latent image, representing the distribution of radioactive substance. CHAPTER 6

MATERIALS AND METHODS

Animals

Pregnant albino mice (Mus musculus Linn.) of the inbred strain

BALB/c were obtained from the Charles River Breeding Laboratories, Inc.

(Wilmington, Mass.). The day of conception indicated by the supplier proved to be only an approximation. The day of gestation on which an injection was given was determined ex post facto from completed slides, by comparing the unknown brain whose pattern of labelling is to be ascribed to a particular day with brains labelled on known days of gestation. The day of conception was considered day zero, the first day of gestation day one, and so forth, following the system used by

Angevine (1965, 1970a). The offspring were designated to show the day of injection and age when sacrificed; EIS-P2!, for example, is a mouse whose mother was injected on the 13th day of pregnancy and sacrificed at four months of age. Gestation in this strain is 18-19 days, but only the tenth through l8th prenatal days plus first and fourth postnatal days are included in this study. Prior to day ten the embryo labels poorly; perhaps the circulation of the placenta does not favor passage.

Certainly the nervous system contains few cells undergoing final divi­ sion. Table 1 lists all of the animals employed in this study, some of which came from other collections as noted in the footnote to the table.

62 63

TABLE 1

ANIMALS EMPLOYED IN THE PRESENT STUDY

Animal Number Day of Injection Day of Sacrifice Series

E10-P7 True E10 7 months Creps-*- 2 Ell-Pl True Ell 1 month Angevine

E12-P6 Early E12 6 months Creps

E12-P1 Late E12 1 month Angevine

E13-P1 True E13 1 month Angevine

E1U-P6 Early Ell+ 6 months Creps

ElU-Pl Late El^t- 1 month Taber Pierce

E15-P9 True E15 9 months Angevine

E16-P2 Early El6 2 months Angevine

E16-P3 Late El6 3 months Creps

E17-P3 True E17 3 months Creps

E18-P1 Early El8 1 month Angevine

E18-P3 Late El8 3 months Creps

P1-P3 Postnatal 1 3 months Creps

PU-P3 Postnatal U 3 months Creps

1. BALB/c X BALB/c

2. BALB/cGn X SJL/J hybrids. Slides from the collection of Dr. Jay B. Angevine, Jr., prepared with the support of USPHS research grant (NB-02853).

3. BALB/cGn X SJL/J hybrids. Slides from the collection of Dr. Elizabeth Taber Pierce, prepared with the support of USPHS research grant (NB-03756). 61*

Injection of Tritiated Thymidine

Pregnant mice were injected intraperitoneally with five

microcuries per gram body weight of tritiated thymidine (l ;u.c/ul;

specific activity 6.3 c/mmole) from a 100 jal syringe and a 27-gauge

needle. Although intravenous injection is preferred (Sidman 1970a),

it is difficult to accomplish in unanesthetized pregnant mice. For

adult mice, intraperitoneal injection is preferable to the subcutaneous

route. With newborns, however, injections were given subcutaneously

since fluids tend to ooze from intraperitoneal injection sites.

It has been shown (Reichard and Estborn 1951) that thymidine is incorporated almost exclusively into DM; thus tritiated thymidine is a tracer of newly formed DM. A single injection is a pulse label

since it is available to dividing cells for a maximum of 30 minutes

after intraperitoneal injection (Skougaard and Stewart 1967).

H3 -thymidine is rapidly broken down or excreted, remaining only in those cells whose DM. was being synthesized while the label was available.

Some labelled thymidine might be stored and utilized at a later time, but the work of Miale and Sidman (1961) suggests that this possibility,. which has never been demonstrated for brain tissue, is unlikely.

The dose of H 3-thymidine used to study the nervous system is greater than that used to study other organs since the nervous system labels poorly at lower doses. The works of Altman (1966), Fujita

(1967), Taber Pierce (1967)5 Fliedner et al. (1968), Trinkaus and Gross

(1961), and many others demonstrate that no toxic effects to the nervous system occur with dosage of 5 juc/gm body weight (Sidman 1970a). 65

Fixation

All animals were sacrificed, as adults (three to nine months of

age). The animals were anesthetized with ether, immobilized on a mouse

dissection board, and fixed by perfusion with 10$ neutral formalin. The liver and heart were exposed and part of the liver removed to allow the perfusate to escape. A 25-gauge needle, attached to a 30-ml syringe by

a piece of tubing, was plunged (slow insertion causes uncontrolled depth of penetration) into the left ventricle and 30 ml of physiological saline perfused to clear the circulatory system of blood and prevent coagulation. A vasodilator (procaine, 0.11$) was included in the saline to improve perfusion. The saline syringe was then replaced with one filled with 30 ml of 10$ neutral formalin. Following formalin perfusion the animal was decapitated and the brain removed from the skull with forceps and iridectomy scissors under a surgical microscope. A successful perfusion was indicated by absence of blood in the superior sagittal sinus; unsuccessfully perfused brains were discarded. The brain was then postfixed in 10$ neutral formalin in the refrigerator overnight, or three to six hours.

Acrolein, a fixative preferred by some (Sidman 1970a), was found to harden tissue excessively and cause difficulty in sectioning.

10$ neutral formalin caused no problems and served adequately. Formalin is reported to desensitize the emulsion slightly (Rogers 1967), but labelling was excellent in this study. Embedding and Sectioning

The fixed brains were washed in water for two hours and then

placed in three changes of 70$ alcohol (one to two hours each), two

changes of 80% alcohol (at least one hour each), and at least four

changes of dioxane (minimum of three hours each). Dioxane, an agent for both dehydrating and clearing tissue (Graupner and Weissberger 1931) is miscible with water, alcohol, hydrocarbons, and paraffin. Tissues can be treated up to a week without harmful effects. Its fumes are toxic, however, so it was stored and used only under the hood.

After the final change of dioxane the brain was infiltrated with

Paraplast at 59° C (three changes of one to two hours each), embedded, and stored at room temperature. Paraplast is a suitable embedment for brain; its melting point is 56-58° C. Polyester wax, recommended by

Sidman, Mottla, and Feder (1961), was not reliable in keeping sections on the slides and sometimes impossible to remove from tissues.

Tissue blocks were trimmed with a razor blade and positioned on a rotary microtome. Coronal sections were cut at 10 u and spread on a water bath 5-10° lower than the melting point of the Paraplast, to which the gelatin had been added (l/2 teaspoon per liter of water). The sections were mounted on clean glass slides and dried in an oven at 59° C for one to three days. The slides were then deparaffinized and hydrated by running them through solutions of xylol, graded alcohols, and distilled water. Staining for Myelin

Slides were stained with Luxol Fast Blue MBSN (E.I. Dupont de

Nemours ana Co., Wilmington, Del.) after drying a few hours. Although

some workers stain sections with Luxol Fast Blue after dipping them in

radiographic emulsion, it proved difficult to stain through the emulsion

even with concentrated solutions and prolonged staining time. Staining

before dipping had no adverse effects on the quality of the autoradio-

graphs.

Slides were placed in 95ajo alcohol for two minutes and then into a

solution of Luxol Fast Blue MBSN (one gm/1000 ml of 95% alcohol plus 5

ml of 10% acetic acid )• Staining dishes were placed in a 59° C oven

overnight. The next day the staining racks were placed in 95% alcohol

to remove excess stain and then washed several times in distilled water.

Sections were differentiated in 0.05% Lithium carbonate for 30 seconds

and afterward in 70% alcohol until the stain is no longer removed

(30-60 seconds). Slides were then placed in distilled water and

examined under the microscope to check differentiation of gray and white matter. If necessary, the slides were run through Lithium car­

bonate and 70% alcohol several times. Fully differentiated sections were rinsed in distilled water and dried. Nuclear staining was done after dipping and developing the slides (see below).

Autoradiography - Dipping

The procedure used in this study was essentially the dipping method of Kopriwa and Leblond (1962), modified according to Coleman

(1965). Kodak NTB-2 emulsion (Eastman Kodak, Rochester, N.Y.) was 68

tested by dipping three clean slides in the emulsion and exposing them

for three days. After development, the level of background was checked

and, if low, the emulsion was diluted l/l with deionized distilled water.

The emulsion was melted in a k0° C water bath and then added to 120 ml , o of water heated to 40 C. Diluted emulsion is sufficiently sensitive,

gives twice as many autoradiographs, and forms a uniform coat less resistant to penetration by stain. The diluted emulsion can be stored in light tight boxes in the refrigerator at 5° C for up to three months.

Once the emulsion was tested and diluted, slides were soaked in distilled water for 30 minutes, placed in holders (Coleman 1965), and set on a hot plate (Uo° C) to dry in the dark room. The emulsion was melted in a 1+0° C water bath in the dark for one hour, then poured into a clean, dry, warm coplin jar. The coplin jar was positioned with grooves to the sides and the slide grips lowered smoothly into the emulsion (sections facing the investigator) and withdrawn swiftly to allow only a thin layer of emulsion to coat the slides. Slides were then returned to the hot plate, dried for two hours in the dark, placed in slide boxes (sections facing Drierite placed at one end), and taped to keep out light. The slide boxes were stored in the cold room at

-15° C for five weeks.

Autoradiography - Development

After five weeks exposure in the cold, slide boxes were removed and allowed to warm to room temperature. The slides were placed in trays in the dark and run through the following at 65-68° F (l6-l8° C): developer (Kodak Dektol, diluted 1/1 with distilled water) for three 69

minutes, distilled water for one minute, fixer (Kodak Acid Fixer) for

ten minutes, and distilled water for one minute.

Nuclear Staining

Slides were rinsed in distilled water and placed in a solution

of Nuclear Fast Red (Matheson, Coleman, and Bell, East Rutherford, N.J.)

made by adding 0.1 gm Nuclear Fast Red to aqueous aluminum sulfate

(5 gm/100 ml of water). After three to five minutes, the slides were

rinsed in distilled water dehydrated in graded alcohols and xylol, and

mounted in Permount.

Analysis of Autoradiograms

After slides were processed and the day of injection estab­

lished, areas of interest were examined microscopically. Nuclei of

radioactive cells could be classified as heavily or lightly labelled

according to the number of overlying silver grains (Angevine 19&5,

1970b). Heavily labelled nuclei are those which synthesized DNA for

the final time within a few hours of injection. Lightly labelled nuclei

synthesized DNA. at that time but divided one or two times before leaving the proliferative zone. Lightly labelled cells have been proven in

Angevine's studies to correspond generally in location and cell type to heavily labelled ones on subsequent days of injection. Some lightly labelled nuclei, however, probably are exceptions and could represent

1) small nuclei, deeply situated, which are highly radioactive but from which few beta particles (tissue course 1.2 u) reach the overlying emulsion; 2) nuclei labelled very early or very late in the bell-shaped 70

curve of the synthetic phase; 3) nuclei whose DNA was donated by a dying

cell (trephocyte) labelled on a previous day; and U) nuclei synthesizing

DNA in areas where uptake or utilization of thymidine was restricted at

the time of injection (Hinds 1967). Despite these qualifications, the

relative number of heavily and lightly labelled cells reliably indicates

the period and pattern of time of origin for all areas of the brain

studied. Differences in intensity of labelling due to variations in

emulsion, development time, and other procedures during slide prepar­

ation can be reconciled by standardizing each slide, i.e., checking

other areas of the brain known to contain heavily labelled neurons on

that day of gestation.

The first step in analyzing the pattern of radioactive neurons

was to examine many slides and determine whether each area under study

contained few, some, or many heavily labelled neurons. This qualitative

step is essential since a reconstruction from only one section may be

atypical of the pattern of labelling on that day (Creps and Angevine

197l). After surveying each nuclear area a representative slide was

selected, viewed at a magnification of 312 (l2.5x oculars and 25x

objective), and the profiles of nuclei of radioactive neurons drawn

to scale in their actual location, using a drawing tube and microscope

(Angevine 1965). Glia were not drawn. At early stages of development it was easy to distinguish glia from neurons by differences in staining and nuclear morphology, but very difficult at late stages in areas where glia and small neurons arise simultaneously. To aid in analysis, approximate nuclear boundaries were drawn.

Although some nuclei are easy to delimit (islands of Calleja), others were arbitrarily delimited with the aid of the atlas by Konig and

Klippel (1963) on the rat brain and that of Sidman, Angevine, and

Taber Pierce (1971) for the mouse. The number of heavily labelled neurons in each nuclear area was determined and graphed. The drawings were traced in ink and reduced photographically. CHAPTER 7

RESULTS

The time of origin (day of final division) for each nucleus studied is illustrated in figs. 2-kG^. These autoradiographic recon­ structions show the actual location of radioactive neurons drawn to scale from "brain sections of adult mice which received a single injection of H 3 -thymidine on known days of development. The number of heavily labelled neurons on each reconstruction was plotted against day of development (figs. V7-5&). The number of lightly labelled neurons was also determined but not plotted. Such a graph would be similar except that more lightly labelled neurons would appear early in development and fewer on late days since most lightly labelled neurons correspond to heavily labelled ones on subsequent days (see Analysis of Autoradiograms).

Although the existence of specific nuclei is generally agreed upon, delimitation of distinct boundaries is often arbitrary and sometimes artificial. To enable a semiquantitative analysis of the results, however, boundaries were drawn with the aid of the mouse atlas by Sidman, Angevine, and Taber Pierce (1971). In addition, the atlas of the rat brain by Konig and Klippel (1963) was employed. The rat brain is similar to the mouse brain except for striking differences in certain areas, for example, the fissural pattern of the cerebellum and

1. For ease of comparison all figures are grouped in the Appendix, p. 117.

72 73 the architectonics of the nucleus of the anterior commissure. Names of

all nuclei and their abbreviations are given in Table 3. Orientation of the five transverse planes (i-V) examined is illustrated in fig. 1. To

facilitate presentation of the results a brief description of each

nucleus will be given. For further anatomical information consult chapter U.

Basal Olfactory Region

Anterior Olfactory Nucleus

The anterior olfactory nucleus is encountered at the rostral- most level lying within the olfactory crus. At this level it completely surrounds the olfactory ventricle and intermediate olfactory tract; its pars medialis, pars dorsalis, and pars lateralis are named relative to these structures. Pars posterior, present only at caudal levels, lies between pars medialis and the ventricle and contains small, compact nuclei. Pars externa lies outside the above subdivisions and also contains small densely packed nuclei. Pars medialis, dorsalis, and lateralis, composed of medium-sized nuclei, are indistinctly separated from each other and adjacent nuclear areas. Pars lateralis is contin­ uous with the polymorph layer of the olfactory tubercle and pyriform cortex; pars medialis grades into the septal area and anterior hippo- campal nucleus (rostral representation of the hippocampus); and pars dorsalis grades into frontal cortex. Pars externa is continuous with pars lateralis at caudal levels. Only pars posterior is distinctly separated from the nucleus accumbens septi lying at its caudal border. 7^

Fiber connections of the anterior olfactory nucleus are presented in

chapter U and details of its structure have been presented by Gurdjian

(1925) for the rat, Young (1936) for the rabbit, Fox (19^-0) for the

cat, and Valverde (1965) for the cat, rat, and mouse.

The anterior olfactory nucleus contains no heavily labelled

neurons in the animals injected on the tenth day of gestation but all

subdivisions contained at least one heavily labelled neuron on the

eleventh day. The pattern of proliferation is distinctive for each

subdivision (figs. 2-7, V7-U8). The peak of proliferation (the day on

which the greatest number of neuron precursors underwent final division)

was earliest in the pars dorsalis (late on E12); slightly later in pars lateralis (E13); still later in pars externa and pars medialis (early on

ElU); and last in pars posterior (late on ElU). Neuron formation was completed by El8 except in pars lateralis and pars posterior, where proliferation continued through the first postnatal day. These late forming neurons are probably derived from the subventricular zone (see

General Observations).

A caudorostral gradient was previously reported (Hinds 1967) in this nucleus but in the present study the only recognizable gradient was in pars medialis of the mouse injected on E17 (the last day of its period of proliferation) where the rostral level contained more heavily labelled neurons than the caudal level. The later period of prolifer­ ation of pars posterior, the most caudal subdivision, makes an over-all caudrorostral gradient unlikely. 75

Nucleus of the Lateral Olfactory Tract

The nucleus of the lateral olfactory tract is a compact

spherical group of neurons in the anterior amygdaloid area, overlying the medial edge of the lateral olfactory tract. Gurdjian (1928) recog­ nized dorsal and ventral subdivisions in the rat, but it is largely homogeneous in the mouse. It receives fibers from the lateral olfactory tract and its output is similar to that of the remainder of the amygdala

(Valverde 1965, Craigie, as reported by Zeman and Innes 1963, Gurdjian

1925, 1928).

The nucleus of the lateral olfactory tract has the extremely early and brief period of origin (E10-E14) characteristic of large cell nuclei (figs. bG, by). Its time of origin is similar to that of the anterior amygdaloid area (E10-13, Hinds 1967) of which it is considered a part.

Olfactory Tubercle

The olfactory tubercle, part of the paleocortex, begins at the caudal end of the anterior olfactory nucleus and extends to the level of the anterior commissure. It is bounded medially by the nucleus of the diagonal band of Broca and laterally by the pyriform cortex. The outer layer, lamina plexiformis, contains principally the fibers of the lateral olfactory tract. The prominent middle layer, lamina pyramidalis, is an undulating sheet of densely packed small and medium-sized neurons.

The inner layer, lamina polymorphica, contains a heterogeneous popu­ lation of medium to large neurons which blend with neighboring nuclei, especially that of the diagonal band posteriorly and the medial septal and anterior olfactory nuclei (pars lateralis) anteriorly. In addition to connections through the lateral olfactory tract the olfactory- tubercle projects primarily to septal area and pyriform cortex (Fox

191+0).

The larger neurons of the polymorph layer have an earlier period of origin than those of the pyramidal layer (fig. 8-33, ^+9)- Although the peak of proliferation occurs only slightly earlier (E13 compared to early ElU), neuron origin begins a day earlier (E10 vs. Ell) and neuron formation is complete by day l8 whereas pyramidal neurons continue to arise through postnatal day one, probably from subventricular cells (see

General Observations). Only the rostral level of the olfactory tubercle was analyzed because it was present inconsistently at caudal levels.

The sequence is "inside-out", just as in all other cortices (Angevine and Sidman 1961, 1962; Hinds and Angevine 1965). Other slides were examined to determine any differences in rostral vs. caudal levels, but no gradient was evident.

Basal Islands

Groups of small celled neurons known as islands of Calleja appear within the olfactory tubercle. Gurdjian (1925) recognized 12 distinct islands in the rat. The largest, the insula magna Calleja, lies outside the tubercle, medial to the nucleus accumbens septi and lateral to the medial septal and diagonal band nuclei. It contains mostly small and a few large neurons. Additional islands, sometimes termed striatal islands, lie within the nucleus accumbens septi. In a recent comprehensive review of the hypothalamus Nauta and Haymaker 77

(1969) state that there are no known connections of the islands of

Calleja but Gurdjian in his 1925 paper described fibers leaving the islands to enter the medial forebrain bundle and tuberculopyriform tract. He speculated that the islands may also contribute fibers to the tracts between the tubercle and the septal, preoptic and hypothalamic areas. Fox (19^0) noted fibers from the islands entering the stria medullaris thalami and Lauer (19^5) mentioned some entering the anterior commissure.

Analysis of the basal islands is difficult because their number and size in any section varies greatly. To facilitate analysis the average number of heavily labelled neurons per island was plotted in fig. 50. Examination of the autoradiographic reconstructions at level

II (figs. 8-22), however, will disclose that all islands do not have an identical pattern of proliferation; in one reconstruction some islands may be labelled and others not.

In general, the islands have a late and prolonged period of origin beginning on E13 for the small ones and late on E12 for the large.

The large neurons of the insula magna originate earlier than its small neurons. Although difficult to determine, the peak of proliferation for the small islands is between the l6th and l8th days of gestation.

The peak for the medial island is El6. Although the neurons of some islands continue to arise through postnatal day four, others cease proliferation sooner.

Although nothing definite can be stated regarding the site of proliferation in this study, Sanides (1957) adduced evidence that the islands come from the subependymal layer. Since in the present study

island neurogenesis continued after the ventricular zone is known to

be absent at least in the adjacent olfactory region (Hinds, 1967), the islands very likely derive from the subventricular zone (see General

Observations).

Preoptic Area

This region, lying between the lamina terminalis and optic chiasm, has been the site of a border dispute. The BNA. (1895)? its

Birmingham Revision (1928), and the current WAP (1955) ignore its existence as such. According to an interpretation of the BNA, based upon studies by His (1888), the telodiencephalic boundary stretches from the velum transversum to the posterior edge of the chiasmatic ridge, dividing the hypothalamus into a telencephalic pars optica, which includes preoptic, supraoptic and tuberal areas, and a diencephalic pars mammillaris. Others (Rudebeck 19^+5? Le Gros Clark 1938) have used derivatives of the transient diencephalic neuromeres as landmarks and shift the ventral boundary marker more rostrally, to the preoptic recess, thus making the preoptic area telencephalic and the remainder of the hypothalamus diencephalic. Christ (1969) and others, however, consider the fundamental longitudinal zones, especially as detailed by

Kuhlenbeck and Haymaker (19^9)5 better landmarks for establishing homologies and place the boundary even more rostrally, extending from the velum transversum dorsally to the torus transversum (corresponding in position to the adult anterior commissure) ventrally, thereby placing 79

the preoptic area within the diencephalon as part of the hypothalamus.

Since "boundaries based on embryonic landmarks are arbitrary and not

universally agreed upon, a possible rationale for classifying the

preoptic area as telencephalic or diencephalic might be provided if

its time of origin is similar to either that of the telencephalic or

diencephalic nuclei which border it (see Discussion). Histologically,

the boundaries of the preoptic area are arbitrary since it "blends

inferiorly with the olfactory tubercle and nucleus of diagonal "band,

laterally with the pyriform cortex, anteriorly and dorsaily with the

septal area, and posteriorly with the supraoptic hypothalamus. The

nuclei of the diagonal band and anterior commissure and part of the

nucleus of the stria terminalis, considered preoptic by some, have "been

included in the present study with the septal nuclei (see Septal Area).

For additional detail consult Gurdjian (1927) or Krieg (1932).

Periventricular Region

The small celled neurons of the periventricular region of the

preoptic area can be divided into three nuclei: preopticus medianus,

preopticus periventricularis, and preopticus suprachiasmatica. The

single median preoptic nucleus is a conspicuous group dorsal and

anterior to the third ventricle. The periventricular nucleus consists

of diffusely arranged cells between the anterior end of the third

ventricle and the medial preoptic nucleus, with which it is continuous.

The supracbiasmatic nucleus, a dense aggregation of round or fusiform neurons dorsal to the optic chiasm and ventrolateral to the third ventricle, is conspicuous in the rat where it has been called the

nucleus ovoidus (Gurdjian 1927).

Of these three nuclei the median originates first, beginning on

E10 and continuing through ElU with the peak of proliferation late on

E12 (figs. 23-^-5) 51)- The periventricular neurons begin to arise at

the same time but their peak is one day later, E13, and proliferation

continues until El6. The suprachiasmatic neurons arise somewhat later

than those of periventricular, beginning on Ell at rostral levels and

peaking early on day 13. Fig. 31 of the rostral level early on El6

illustrates the difference in time of origin in these three nuclei;

there are no heavily labelled neurons in the median nucleus, few in

the periventricular nucleus, and several in the suprachiasmatic nucleus.

This sequence in time of final division demonstrates a dorsoventral

gradient similar- to that found in the periventricular region of the

hypothalamus (Angevine 1970c). Only slight differences were observed

at rostral vs. caudal levels.

Medial Preoptic Nucleus

This nucleus contains medium-sized neurons and lies lateral

to the periventricular region, medial to the nucleus of the stria

terminalis and lateral preoptic nucleus, dorsal to the nucleus of the

diagonal "band, and ventral to the anterior commissure and septal area.

It has the longest period of proliferation of all preoptic nuclei

(figs. 23-^45, 52). A moderate number of neurons undergo final division on E10 with an increasing number up to day 13. The number of prolifer­

ating neurons then drops off late on the l8th day. Although no gradient along the rostrocaudal axis was noticed, an "outside-in" gradient was

apparent relative to the lateral preoptic nucleus (see Lateral Preoptic

Region).

The period of proliferation of medial preoptic neurons is

distinctly longer than that of periventricular neurons just described,

although the onset of neurogenesis is the same for both. It is clear

that many of the medial preoptic neurons, or their cell nuclei, must

have migrated or otherwise transposed with periventricular ones in an

"inside-out" pattern of proliferation.

Lateral Preoptic Region

The lateral preoptic region contains mainly fibers of the medial forebrain bundle, the interstitial neurons of which form the so-called lateral preoptic nucleus. These medium-sized neurons extend laterally from the medial septal nucleus to the pyriform cortex, lying ventral to the bed nucleus of the stria terminalis and dorsal to the nucleus of the diagonal band and magnocellular preoptic nucleus.

The pattern of proliferation is similar at first to that of the medial preoptic nucleus, beginning on E10 with the peak on E13 (figs.

23-^5, 52). After E13, however, neuron production drops rapidly and no neurons arise after El6, two days sooner than for the medial nucleus.

In terms of cessation, therefore, this sequence in time of origin is

"outside-in" since the last cells to arise lie closest to the ventricle.

This gradient, however, is not as dramatic as the septal gradient (see above) and furthermore must exclude the periventricular region where neuron proliferation ceases by E17 (see above). Magnocellular Preoptic Nucleus

Although classified as preoptic, most consider the magnocellular

nucleus a subdivision of the nucleus of the diagonal band (Young 1936).

Its large neurons extend from the olfactory tubercle to the lateral

preoptic nucleus and lie at the junction of the fibers of the medial

forebrain bundle and diagonal band. Laterally this nucleus terminates

in two cell cords, a dorsal one which blends with the globus pallidus

and a ventral one which grades into the anterior amygdaloid area.

The magnocellular nucleus has the shortest period of origin of

the preoptic nuclei (figs. 23-^5, 53). Neurogenesis is well underway

by E10 and increases through E12, after which it drops precipitously to

a close on E13. This brief and early span of origin has been demon­

strated for other large neurons (see Discussion). In this case, it is

even shorter than that for the nucleus of the diagonal band, of which it

is considered a subdivision. The pattern of final division is similar

at all levels.

Septal Area

The septal area is a subdivision of every vertebrate brain, as

Johnston and Herrick clearly showed in their many contributions early

in this century, but because of the complicated development of the

mammalian brain it is hard to determine homologies to parts of the massive and clearly recognizable septal area of other vertebrates

(Loo 19^1). Consequently, a cumbersome terminology has developed.

The part dorsal and posterior to the anterior commissure is generally acknowledged as septal, but that anterior to the commissure is not. The correspondence of certain parts of the precommissural

septum (Crosby and Humphrey 19^-1) to components of the septal area of

other vertebrates has been questioned. Hence, such names have been

provided for this precommissural region as subcallosal gyrus of Zucker-

kandl (l888), paraterminal body, and later precommissural body, by

Elliot Smith (1896, 1903), and parolfactory area by Johnston (1913).

The septal terminology generally used today and followed in this paper

is that of Edinger (1908-1911) and Herrick (19IO).

Although it forms the medial wall of the cerebral hemisphere,

the septal area is not cortical (laminar) in organization, but nuclear.

In primates the supracommissural septum, between the corpus callosum and

columns of the fornix, consists principally of septum pellucidum and

contains few nerve cells. In some other mammals (including the mouse) this part contains nuclei and its posterior margin is clearly indicated by the relatively large hippocampal commissure, or ventral psalterium.

The precommissural septal area is continuous with the anterior olfactory nucleus and anterior hippocampal nucleus, the olfactory tubercle, and the preoptic area. The nucleus of the diagonal band of Broca and interstitial nuclei of the stria terminalis and anterior commissure are considered septal components by some investigators because of their close relationship with other septal nuclei (Andy and Stephan 1966) and are analyzed as such in this study. The connections of the septal area are mainly through the medial forebrain bundle (Millhouse 1969), the fornix (Wauta 1956), the diagonal band (Price and Powell 1970b), and the stria terminalis (Gurdjian 1925). 84

Medial Septal Nucleus

This nucleus, comparatively small in the mouse contains large

multiform neurons and occupies the medial region of the septal area

extending from the anterior hippocampal nucleus to the level of the

anterior commissure. It is continuous with the nucleus of the diagonal

band.

The medial nucleus has the shortest, earliest period of origin

of the septal nuclei, from E10 through Ell+ (figs. 8-22, 2b, 5^) • The

majority of the neurons arise late on E12 or on E13. More laterally

located septal neurons continue to arise until El8, illustrating an

"outside-in" gradient of proliferation (see below). The time of origin

of the medial nucleus is slightly earlier than that of the caudally

located triangular and fimbrial nuclei, demonstrating a rostrocaudal

gradient within the septal area (see below).

Triangular Septal Nucleus

This nucleus contains small and medium-sized neurons and

occupies the triangular area between the columns of the fornix and

anterior commissure. It lies posterior to the medial septal nucleus, wholly within the supracommissural part of the septum.

Few neurons arise on E10 and 11; most arise on E12 and 13 with a few originating through E15 (figs. 23-^-5, 55). This span of origin is slightly longer than that of the medial septal nucleus (see above) and essentially identical to that of the fimbrial nucleus (see below).

Within the triangular nucleus, the peak of proliferation is slightly 85

later at the caudal level (E13 compared to late El^>, possibly indicating

a rostrocaudal gradient.

Fimbrial Septal Nucleus

This nucleus, composed of medium-sized neurons, lies lateral

to the columns of the fornix and is really the interstitial nucleus

of the ventral hippocampal commissure. It lies wholly within the

supracommissural septum, medial to the lateral septal nucleus.

Its pattern of neuron origin is essentially identical to that

of the triangular nucleus (figs. 23-^+5? 55). Few neurons undergo final

division on E10 and 11; most arise on E12 and 13- Proliferation rapidly

drops after E13 and ceases completely by El6. This period of origin,

later than that of the medial septal nucleus and earlier than that of

the lateral, suggests that the first cells to proliferate lie farthest

from the lateral ventricle (their probable site of origin) in an

"outside-in" sequence (see below). The peak of proliferation occurs on E13 at the caudal level and slightly earlier at the rostral level

(late on E12), indicating the same slight rostrocaudal gradient seen in the triangular nucleus (see above).

Lateral Septal Nucleus

The lateral septal nucleus contains medium to large neurons and extends almost the entire length of the septal area. • In the precommissural septal area it lies lateral to the medial septal nucleus, medial to the lateral ventricle, and dorsal to the nucleus accumbens septi. In the supracommissural area it lies lateral to the triangular 86

and fimbrial nuclei and dorsal to the interstitial nucleus of the stria

terminalis.

This nucleus has a long, late period of origin compared to more

medial septal nuclei (fig. 5^). Few neurons undergo final division on

E10 and 11; most arise on E12 and 13, as in the other septal nuclei (see

above). Neuron proliferation drops off slowly, however, after ElU and

continues through El8, two days longer than the triangular and fimbrial

nuclei and three days longer than the medial septal nucleus. This

sequence, medial, triangular and fimbrial, and then lateral, clearly

shows an "outside-in" gradient, most evident in fig. 16 where the last

neurons to undergo final division lie near the ependyma of the lateral ventricle. Furthermore, the late forming cells lie near the base of the ventricle (figs. 16-19), indicating this angle as the probable site of origin of these neurons. A rostrocaudal gradient is indicated by the slightly earlier peak of proliferation (E12) at the rostral level (figs.

8-22) compared to that (E13) at caudal levels (figs. 23-^-5), an observa­ tion similar to that found in the fimbrial and triangular nuclei (see above).

Nucleus Accumbens Septi

This nucleus lies wholly within the precommissural septum and surrounds the anterior limb of the anterior commissure. It is ventral to the base of the lateral ventricle and lateral to the large island of

Calleja and nucleus of the diagonal band. It is continuous with the polymorph layer of the olfactory tubercle ventrally and the caudate nucleus laterally. Some people consider it an extension of the caudate into the septal area (Young 1936, Fox 19^-0, Lauer 19^5, and others).

This nucleus has the longest period of origin of all forebrain nuclei studied (figs. 8-22, 56). Its neurons begin final division on

ElO and most arise from E15-17 but a few continue to divide through the fourth postnatal day. This period of origin is much longer and later than that of the nucleus of the stria terminalis (E10-18) just caudal to it, but only slightly later than that of the pars posterior of the anterior olfactory nucleus (Ell-Pl) immediately rostral. Although the site of origin was not studied, the protracted time of origin strongly implicates the subventricular zone in production of the late forming accumbens neurons (see General Observations). The time of origin of the neurons of the basal ganglia, although not studied in detail, was examined to determine any similarity with nucleus accumbens. Labelled neurons, especially in the head of the caudate, were present, through the first postnatal day, suggesting a close developmental relationship with nucleus accumbens.

Nucleus of the Diagonal Band of Broca

These large interstitial neurons are found throughout the course of the band, which extends from the anterior amygdaloid area to the medial wall of the hemisphere. Those of the ascending limb lie within the precommissural septum, medial to the nucleus accumbens and insula magna. Those of the horizontal limb pass ventral to the lateral preoptic region and dorsal to the olfactory tubercle, where the cells mingle with those of the polymorph layer. Its largest, most lateral 88

neurons are usually considered preoptic and called the magnocellular

preoptic nucleus, as in the present paper (see above). Some diagonal

band neurons provide efferent fibers to the olfactory bulb for modu­

lation of olfactory input (Price and Powell 1970a). Other connections

of the nucleus are discussed by Price and Powell (1970b).

Few diagonal band neurons arise on E10. Increasing numbers

arise through E13, after which neuron formation drops rapidly, ceasing

on E15 at the caudal level, early on El6 at the middle level, and late

on El6 at the rostral level (figs. 8-U5, 53). Although the findings

would appear to suggest a slight caudorostral gradient, the temporal

pattern of origin is virtually identical at all levels. The staggered end points are probably due instead to the slightly greater numbers of neurons at more rostral levels. The period of origin is distinctly different from those of the three structures with which it is continuous.

It is longer than those of the medial septal and magnocellular preoptic nuclei, but shorter than that of the polymorph layer of the olfactory tubercle.

Interstitial Nucleus of the Stria Terminalis

This nucleus has a septal part dorsal to the anterior commissure and ventral to the lateral septal nucleus and a preoptic part dorsal to the lateral preoptic region. It is generally considered a septal nucleus (Andy and Stephan 1966 , and the present author) even though it is continuous with the medial preoptic nucleus (Wauta and Haymaker

1969). Its medium-sized neurons mingle with those of nucleus accumbens rostral to the anterior commissure. The fiber connections of the stria 89 terminalis are given by Gurdjian (1925), Berkelbach van der Sprenkel

(1926), and Nauta and Haymaker (1969).

The period of origin is shorter and earlier than that of nucleus

accumbens (fig. 5&), beginning on E10 for all levels. Although the peak is late on E12 at the rostral level and E13 at the caudal, there is no clear-cut rostrocaudal gradient, since the end points are El8 for the rostral level and E17 for the caudal (figs. 23-^5). This data sheds no light on the regional affiliations of this nucleus since the time of

origin is identical to that of the lateral septal nucleus on the one hand (fig. 5^) and medial preoptic nucleus on the other (fig. 52).

Nucleus of the Anterior Commissure

These small to medium-sized, densely packed neurons are conspicuous in the mouse; they lie dorsal to the anterior commissure, just medial to the nucleus of the stria terminalis. This nucleus is considered septal by some (Fox 19^-0, Andy and Stephan 1966, and the present author). For a description of its connections consult Gurdjian

(1925) or Brodal (19U8).

The period of origin of this nucleus is very brief and early, offering a strong contrast to nuclei with similar-sized neurons (fig.

50). Although a few neurons arise on E10-12, most form on E13, and all have arisen by the end of the l6th day of gestation (figs. 3^-^-5) •

Summary of Observations

Origin of basal forebrain neurons begins at a very early point in brain development, on the tenth embryonic day in almost every area 90

examined. Exceptions are the anterior olfactory nucleus and pyramidal

layer of the olfactory tubercle, whose neurons begin to arise a day

later, and the islands of Calleja, for which time of origin begins two

to three days later. All neurons have arisen by the end of gestation

except for the lateral and posterior parts of the anterior olfactory

nucleus, the pyramidal layer of the olfactory tubercle, the islands of

Calleja, and the nucleus accumbens. Postnatal genesis of these neurons

implies that their site of origin is the subventricular zone. This

zone is prominent at birth in the mouse, especially around the anterior

horn of the lateral ventricle (Smart 1961), whereas the ventricular zone

is absent, at least in the nearby olfactory region (Hinds 1967), after

the 17th prenatal day.

The findings fortify certain general principles of neuronal

development, recognized by early workers but reaffirmed with special

clarity in time of origin studies. Once again, large neurons (medial

septal, magnocellular preoptic, and nucleus of the lateral olfactory

tract) generally arise earlier than small ones (nucleus accumbens and

islands of Calleja), but of course exceptions occur (early origin of

nucleus of the anterior commissure and late origin of nucleus of the

diagonal band). In some cases, sharp differences in time of origin

accord well with nuclear boundaries (nucleus accumbens and pars posterior of the anterior olfactory nucleus). In other cases, however, neighboring nuclei have identical patterns of proliferation (triangular

and fimbrial, for one example, and lateral septal, stria terminalis,

and medial preoptic nuclei for another). A striking mediolateral gradient seen within the septal area is

"outside-in" relative to the lateral ventricle (the probable site of origin). A dorsoventral gradient in the periventricular preoptic region is similar to that reported in the hypothalamus (Angevine 1970c in mouse, Ifft 1972 in rat, Keyser 1972 in the hamster). In addition, the medial preoptic nucleus has a longer time span of origin than the lateral, suggesting a lateromedial gradient ("outside-in" relative to the third ventricle) similar to that noted in the hypothalamus (Angevine

1970c) and in the preopticohypothalamic region of the rat (ifft 1972), and hamster (Keyser 1972). An opposite gradient is apparent, however, when the medial preoptic and periventricular nuclei are compared. Late forming neurons of the medial preoptic nucleus apparently transpose their perikarya with those of the previously formed periventricular neurons in an "inside-out" sequence. The olfactory tubercle also arises in an "inside-out" sequence (polymorph layer preceding pyramidal layer) indicating migrations of neuronal perikarya past previously formed ones, as in other cortices (Angevine and Sidman 1961, 19&2; Hinds and Angevine

1965). The only rostrocaudal gradient observed was in the septal area where rostrally located medial septal nucleus had the earliest time of origin and the rostral level of the lateral, triangular and fimbrial nuclei arose before the caudal level. Wo other gradients were detected. CHAPTER 8

DISCUSSION

Significance of Results

Plotting the location of radioactive neurons in brain sections

of adult mice given a single injection of H^-thymidine on a known day

of development is a reliable method for establishing the time of neuron

origin (the day when the neuronal precursor underwent final division).

Small errors, resulting from difficulty in judging lightly vs. heavily labelled nuclei, or differentiating small neurons from glia, are mini­ mized if procedures for processing slides are standardized and a series of animals is injected at closely spaced intervals (see Materials and

Methods). Since there is apparently little individual variation in time of origin within the same species, the investigator can make generalized statements from examination of only one or two brains for each stage of development, provided such examination is systematic and exhaustive.

Angevine (1965) states that "...plots made from litter mates injected and processed in identical fashion could be superimposed to show remark­ able similarity in number and placement of labeled neurons" (p. 28).

Comparison of Results with Other Studies

Autoradiographic studies of time of origin of basal forebrain neurons in the mouse have been restricted in scope and been published only in preliminary form (Sidman and Angevine 1962, Hinds 1967* Creps and Angevine 1971). Table 2 provides a comparison of the present

92 TABLE 2

COMPARISON OF RESULTS WITH OTHER STUDIES

Nucleus Present Study Other Studies (Hinds 19673 Mouse)

Total Period Peak Total Period Peak

Anterior Olfactory Pars Medialis Ell-17 Late 12-15 Ell-15?" E12-15 Pars Dorsalis Ell-17 Late 12-lU E12-18 E12-17 Pars Externa Ell-17 Early lU E12-16 E12-1U Pars Lateralis Ell-Pl E13 E12-18!" E12-18 Pars Posterior Ell-Pl E13-l^ Ell-15 E12-1U

Lateral Olfactory Tract E10-14 Late 12-13 Ell-13 E12

Olfactory Tubercle Polymorph Layer E10-17, Late 12-13 Ell-15 E12-15 Pyramidal Layer Ell-Pl Early 14 Ell-15 E12-15

Islands of Calleja E13-p^ E16-18 E13-18 E15-18 Insula Magna Late 12-PU E16-18

Septal Area Medial ElO-lU. Late 12-13 ElO-lUj Ell-13 Lateral E10-18 Late 12-13 E10-163 Ell-15 Triangular ElO-15^ E12-13 E10-14 Ell-13 Fimbrial E10-15 E12-13 ElO-lU Ell-13 Accumbens E10-P4 E13-18 E12-18 E15-18 Stria Terminalis E10-17 Late 12-13 E10-15 E12-13 Diagonal Band E10-16 E13 ElO-lU E12-13 Anterior Commissure EIO-I6 E13 TABLE 2 (continued)

Nucleus Present Study Other Studies (ifft 1972, Rat)

Total Period Peak Total Period Peak

Preoptic Area Median ELO-LU E12-13 Periventricular E10-163 E12-13 E12-18 El6 Suprachiasmatic E10-16 E13-15 E12-18. EI6 Medial E10-18 Late 12-13 E12-18, EI6 Lateral E10-164 Late 12-13 E12-17 EI6 Magnocellular E10-13 E12

1. Caudorostral gradient 2. Rostrocaudal gradient 3. Mediolateral gradient Lateromedial gradient 95

results with those of other studies in the mouse telencephalon (Hinds

1967) and similar studies in the rat preoptic area (ifft 1972).

The anterior olfactory neurons begin to arise one day later

than most septal and preoptic neurons. The times of origin offer an

exception to the generalization that larger neurons (pars medialis,

dorsalis and lateralis) arise earlier than small neurons (pars posterior

and pars externa). According to the present findings the medial, dorsal, and external subdivisions of the anterior olfactory nucleus

arise concomitantly (Ell-17), with pars dorsalis having a slightly earlier peak of proliferation. Pars lateralis and posterior begin to

arise similarly, but continue through PI, pars posterior reaching its

maximum proliferation before pars lateralis. According to Hinds (1967), however, pars lateralis and dorsalis have identical periods of prolif­ eration (E12-18), later than those of pars externa (E12-16), medialis

(E12-15), and posterior (also E12-15). He reported a caudorostral gradient within the nucleus, similar to a gradient in differentiation reported in the human (Humphrey 1963). Although a slight gradient was noted in this study, the time of origin of pars posterior (Ell-Pl), much longer than that reported by Hinds, rules out an over-all caudorostral gradient.

That the time of origin of the olfactory tubercle follows an

'1 nside-out" gradient was not recognized by Hinds (1967). The large neurons of the polymorph layer arise earlier (E10-17) than those of the pyramidal layer (Ell-Pl). A caudorostral gradient in differentiation in 96

the human tubercle (Humphrey 1966) was not corroborated by a gradient in

time origin in the mouse.

The time of origin of septal neurons generally agrees with that

in other studies (Sidman and Angevine 1962, Hinds 1967) except that

Hinds noted a caudorostral gradient in the medial nucleus, whereas the

present study showed a rostrocaudal gradient for the whole area. Medial

septal neurons arose earliest (E10-lU), triangular and fimbrial longer

(E10-15), especially at the caudal level, and the caudal neurons of

the lateral nucleus later (E10-18). A mediolateral gradient (but

"outside-in" because of the location of the area with respect to the

ventricle) illustrated by the present results, was noted in previous

studies. The periods of origin of the nuclei of the stria terminalis

(E10-17), diagonal band (E10-16), and lateral olfactory tract (ElO-lU),

are consistent with those reported by Hinds. Interstitial neurons of

the anterior commissure, not previously studied, arise from E10 to El6,

earlier than the constituents of most nuclei with similar-sized neurons.

The end points of neuron origin in nucleus accumbens (E10-PU) and

islands of Calleja (E12-P^) were not previously determined.

Preoptic neuron origin has not been studied in the mouse but

has been examined in the rat (ifft 1972). A lateromedial gradient

("outside-in") was consistently noted; in the present study medial

neurons arose from E10 to El8 and lateral ones E10-16. A mediolateral gradient ("inside-out") in neuron origin of medial vs. periventricular region reported in the present study was not observed in the rat. On the other hand, a possible dorsoventral gradient noted in the rat was 97 clearly shown in the mouse (median, ElO-l^l-; periventricular, E10-16,

suprachiasmatic, E10-16). The magnocellular nucleus, not studied

previously, has the primacy of origin characteristic of large-celled

nuclei (E10-13).

Even though some inconsistencies appear, especially in the

anterior olfactory nucleus, the results essentially agree with earlier

reports. The results reaffirm the principle of early origin of large

neurons (Hinds 19&7, 1968a; Sidman 1970a, Angevine 1970a), with some

exceptions. In addition, previously unexplored areas have been

investigated.

Postnatal Neurogenesis

A number of instances of postnatal neurogenesis are now known

in rodents and other precocial mammals. These events are characterized

by unusual features of various kinds, but are all of great interest,

some to the neuroanatomist and others to the behaviorist.

The source of neurons in the perinatal and postnatal mouse is

the subventricular zone, composed of cells which have migrated from the ventricular zone and accumulated between it and the intermediate zone.

Unlike neuroblasts, as currently defined (Boulder Committee 1970), subventricular cells have proliferative capability. Although both neurons and glia are produced, it is not known whether the cells are

'truly indifferent (Schaper 1897) or whether there are separate neuronal and glial precursors. The subventricular zone is presumably present throughout the nervous system at some stage in development (Sidman

1970b), but is especially prominent in the roof of the fourth ventricle prenatally (Sidman 1970b) and around the anterior horns of the lateral ventricles at birth (Smart 1961). The subventricular zone gives rise to

granule cells of the olfactory bulb (Hinds 1968b, Altman 1969b) and the dentate gyrus of the hippocampal formation (Angevine 1965, Altman 1969a).

In the brainstem, subventricular cells from the rhombic lip give rise to two transient structures, the well-known external granular layer of the cerebellum and the pontobulbar body. The external granular layer produces cerebellar granule cells (Sidman and Miale i960). The ponto­ bulbar body produces neurons of the cochlear nuclei (Taber Pierce 1967) certain pontine nuclei, and the inferior olivary complex (Taber 1963,

Taber Pierce 1966). Another transient aggregate of subventricular cells, the human corpus gangliothalamicus, derives from the ganglionic eminence and produces some neurons of the pulvinar (Rakic and Sidman 1969)•

In the present study, postnatal neurogenesis was noted in the pyramidal layer of the olfactory tubercle and the lateral and posterior parts of the anterior olfactory nucleus (see above), but was especially prominent in the islands of Calleja (E12-P^), and the nucleus accumbens

(ElO-PU). Although their site of origin was not studied, this prolonged time of origin suggests that the late-forming neurons are derived from the subventricular zone, prominent at this time in the walls of the lateral ventricles in the mouse (see above). The ventricular zone is absent after the 17th prenatal day, at least in the nearby olfactory ventricle (Hinds 1967). The findings support the suggestion (Sanides

1957) that the islands of Calleja derive from the subventricular zone.

Postnatal neurogenesis in the nucleus accumbens of the rat has been reported (Das and Altman 1970); in a later review (Altman 1970) the site

of origin is postulated to be the subventricular zone on the lateral

walls of the lateral ventricle.

Since it is well known that the environment can affect the

development of behavioral responses (Levine 1962, 1966) many investi­

gators have looked for changes in brain morphology to account for this

behavioral plasticity. It has been shown that rearing animals in

enriched, compared to deprived, environments affects the concentration

of cholinesterase in some brain areas (Bennett et al. I96U) and the

width of the cerebral cortex (Diamond, Krech, and Rosenzweig I96U,

Diamond et al. 1966). It was assumed that environmental conditions did

not alter the number of neurons, since most neurons arise prenatally

when the environment is constant. Recently, changes in dendritic

arborization due to experience have been shown. According to Valverde

(1967) mice raised in darkness have fewer apical dendritic thorns on

cells of their visual cortex compared to mice raised with light stimu­

lation. Conversely, Holloway (1966) showed that stellate cells of the

visual cortex of rats reared in enriched environments have a greater number of dendritic arborizations.

Earlier, Altman and Das (196^) observed an increase in the number of labelled cells in the cerebral cortex of rats raised in enriched environments compared to controls. This increment was originally interpreted as an increase in glial production but is now considered to include some neuronal precursors (Altman 1970). In another study (Altman, Das, and Anderson 1968) handled rats had smaller 100

brains compared to controls. There were no changes in the perikaryon/

neuropil ratio, but the number of labelled cells in the external

granular layer of the cerebellum and in the dentate gyrus of the

hippocampus was increased. Altman (1970) believes that the early

postnatal environment affects brain structure primarily by altering the number of microneurons (interneurons). He states that:

The postnatal interdigitation of microneurons into the frame­ work provided by the morphogenetically determined and prenatally formed macroneurons represents the modification of the structural organization of the brain, its fine-wiring, which is sensitive to environmental influences and is the substrate of plasticity during its early development (p. 237).

Postnatal neurogenesis, however, is not ubiquitous. It has been demonstrated only in precocial mammals in a relatively small number of neuronal aggregates. One might expect plastic changes to occur in the cerebral cortex, but few cortical neurons, if any, arise postnatally, except possibly in the human (Brun 1965). Alterations in the postnatal proliferation of microneurons obviously is only one of several possible mechanisms underlying behavioral plasticity, especially since other studies (see above) demonstrate changes in dendritic and synaptic elabo­ ration. Biochemical changes, for example in RNA and protein synthesis, have also been implicated in long-term and short-term memory, and there are probably many other yet unexplored mechanisms.

Gradients

It is now recognized (see above) that in many prospective brain regions time of origin proceeds in a sequential manner along the wall of the neural tube. These gradients are typically described relative to three axes of the tube, radially from the lumen (lateromedial or 101

mediolateral), longitudinally (rostrocaudal or caudorostral), and

vertically (dorsoventral or ventrodorsal). The latter two suggest

a wave of attainment of final DNA synthesis sweeping over the neural

tube, but the cause and significance are unknown.

A caudorostral gradient (neurons arise at caudal levels before

those at rostral levels) has been reported in the nucleus gracilis

(Taber 1963), parts of the trigeminal complex (Taber Pierce 1970),

thalamus (Angevine 1970a), and the cerebral cortex (Angevine and Sidman

I96I, 1962). An opposite sequence (rostrocaudal gradient) occurs in the

hypothalamus (Angevine 1970c), and according to the present study, in

the septal area. The septal gradient could represent accretion of

neurons on the caudal surface of the septal area as the ventricle

receded in the caudal direction. Viewed from the sagittal surface the

gradient would be classified as "outside-in".

Full understanding will probably not come even with the complete

cataloguing of time of origin in every area of the brain in a single

species. Such a catalogue will certainly be valuable and is nearly

complete in the mouse, especially in the diencephalon (see below). But

comprehension must await consolidation of these findings with other data

on early neurogenesis, much of which, for example, the morphology of the

neural tube as visualized by scanning electron microscopy, has just

been, or is about to be, attained.

Just as early investigators tried to correlate the presence of neuromeres, and later migration bands, to centers of intense prolifera­ tion (Bergquist and Kallen 195*0? attempts understandably have been made 102

to relate gradients in time of origin to other events in neurogenesis.

Angevine (1970a) suggested a relationship between a caudorostral

gradient in the thalamus and the presence of the cranial tip of the

notochord at its posterior boundary. He also discussed the possibility

that thalamic gradients represent revectored gradients known to exist

within the neuromeres when the latter are reorganized into longitudinal

diencephalic zones (Coggeshall 196^). Keyser (1972), however, considers

such revectoring unlikely since, at least in the Chinese hamster,

gradients in matrix layer development similar in direction to those

inferred by Angevine appeared before the neuromeres underwent reorien­ tation.

It has been demonstrated in some brain regions that gradients in time of origin correlate with subsequent times of nuclear recognition

(see Relationship of Time of Origin to Other Events in Neurogenesis) but the time interval varies greatly for individual nuclei. Keyser (1972) showed that, despite a caudorostral gradient of differentiation within the hypothalamus, no gradients in time of origin along the rostrocaudal axis were apparent. That all events in neurogenesis need not occur along an identical gradient is evident since time of origin and differ­ entiation within the brain generally follow a caudorostral trend, while neuromeres (also related to proliferation) usually form in a rostro­ caudal sequence (Bergquist and Kallen 195*0- Thus, as stated above, more information is necessary before the significance of these gradients in time of origin and their relationship to other neurogenetic events are apparent. 103

Another obvious gradient noted in the thalamus is ventrodorsal,

neurons arising at the prospective ventral thalamic region of the

diencephalic ventricular zone before they do at more dorsal latitudes

(Angevine 1970a). An opposite dorsoventral gradient was seen in the

hypothalamus (Angevine 1970c) and, from the present report, the preoptic

area. That these two gradients diverge in opposite directions from the

region of the hypothalamic sulcus endows the sulcus with more than

purely landmark significance and may provide helpful clues to the

genesis of this and other sulci. But why this region of the ventricular

zone should be the first to achieve final DM synthesis escapes explana­

tion by the present findings.

Gradients along the radial axis have been classified as

"outside-in" if the first neurons to arise lie, at maturity, farthest

from their site of origin and "inside-out" if they settle nearest their

site of origin. An "outside-in" sequence suggests that neurons simply

add to the surface, by appositional growth and possible outward

displacement, of a mass of cells previously formed, whereas an "inside-

out" sequence suggests that late forming neurons, or at least their

nuclei, move past, or in some manner transpose with, those previously

formed. The latter gradient was initially interpreted as evidence for

actual migration of neurons (Angevine and Sidman 1962) but now that it

is known that neuronal perikarya can move along a neuronal process without retracting the foot attachment (Morest 1968a), the transposition of nuclei implicit from autoradiography may simply represent perikaryal

shifting rather than actual migration (Angevine 1970b). IOU

Although apparently of opposite sign, the raediolateral gradient

in the septal area and lateromedial gradient in preoptic area (see

Results) are both "outside-in", the septal gradient relative to the

lateral ventricle and the preoptic to the third ventricle (their

probably sites of origin). An "inside-out" gradient also occurs in

the periventricular and medial preoptic regions. These findings are

worth consideration, since it has been suggested (Angevine and Sidman

1962, Angevine 1970a) that the "outside-in" sequence characterizes

nuclear areas whereas the "inside-out" gradient characterizes laminar

structures. "inside-out" gradients in time of origin have been reported

in substantia nigra (Hanaway, McConnell, and Netsky 1971), optic tectum

of the chick (Fujita 196U), olfactory bulb (Hinds 1968a), most of the

hippocampal formation (Angevine 1965), and cerebral cortex (Angevine and

Sidman 1961, I962). These areas all have a laminar or columnar, rather than nuclear, arrangement. The periventricular preoptic region is also laminar; here again, it develops in "inside-out" sequence relative to the medial preoptic nucleus (see above). "Outside-in" gradients have been noted recently in the hypothalamus (Angevine 1970c, Ifft 1972,

Keyser 1972) and the preoptic and septal areas (the present study).

These areas also have a nuclear organization, rather than laminar.

The one possible exception to the above apparently valid gener­ alization is the granular layer of the dentate gyrus, which arises in

"outside-in" sequence (Angevine 1965). These radial gradients are classified relative to their site of origin, however, as well as to one of three axes of the early neural tube. At late stages when this 105

"outside-in' gradient is observed some of the granule cells are arising

from proliferative precursors in the layer itself, which in turn are

apparently coming from the displaced subventricular zone of the deep

root of the alveus adjacent to the fimbria now on the outer, not inner

surface of the brain. So classification is complicated, to say the least. But if the neurons in the hilus of the dentate gyrus are consid­

ered part of that gyrus rather than the hippocampus then the sequence in the dentate gyrus as a whole is "inside-out", as in other cortical areas. Therefore, in the many areas studied thus far, nuclear areas

arise "outside-in" whereas laminar areas arise "inside-out" .

Time of Origin Related to Other Events in Neurogenesis

Time of origin is only one of several stages in neuronal devel­ opment, which also includes, among many other events, neuroblast differentiation, migration to adult location, growth and maturation into adult neuronal type, and attainment of functional competence.

Usually within 2k hours the young neuroblast leaves the ventricular zone and begins to migrate to its final location. This may be an actual migration of the entire cell, which is a classical and widely held view and should not be lightly cast aside. Yet where new findings abound, it is the cell nucleus which migrates; the perikaryon enters the mantle

(intermediate) layer by sliding along a neuronal process, the process being withdrawn sometime later (Morest 1969a, 1969b). There are many variations in details of this migratory process (Morest 1968a, 1968b,

1969a, 1969b, 1970), the mechanisms and controlling factors of which are still undetermined. 106

The sequence of events in neurogenesis may vary depending on

the region of nervous system being studied (Morest 1970). Migration

may precede proliferation; some neuronal precursors migrate to a

secondary location (subventricular zone) and continue to proliferate

(see Postnatal Neurogenesis). Neuroblast differentiation, most reliably

characterized by the presence of an axon (Lyser 1964, Stensaas 1967b,

Morest 1970), can occur before, during, or after migration. If in

earlier stages some neuroblasts can be identified by the presence of

neurofilaments (Sechrist 1969), then differentiation may even occur

before mitosis. Neuroblast differentiation implies that a cell is no

longer undifferentiated and will become a mature neuron rather than a

glial cell. It does not mean that this neuroblast is completely deter­

mined as a specific neuron type which will lie within a specific nuclear

field.

Bergquist and Kallen (195^) implied that neuroblasts form in the

matrix layer and move as a group into the mantle layer where they form

a recognizable adult nucleus; therefore, in this sense, neuroblasts were thought to be predetermined. Coggeshall (1964), however, found that in the developing rat diencephalon all nuclei form in their adult location and that neuroblasts migrate as units, not in groups, and take on no characteristics of their adult cell type until they have reached their final location.

A nuclear area is recognizable only after several neurons reach their final location and differentiate into their specific cell types. 107

The interval of time between proliferation and recognizable nuclear

aggregation almost certainly must be a function of the rate and distance

of migration and possibly of the degree of fiber development at the

final location (see below), hence, there is great variability in this

interval for different nuclear areas. In the olfactory bulb (Hinds

1968b) the sequence in time of origin of the different neuron types

parallels their sequence in differentiation. In the cerebellum, however,

Purkinje cells arise earlier than granule cells, but differentiate later

(Rakic and Sidman 1970). In the thalamus (Angevine 1970a), gradients of

proliferation parallel similar gradients in differentiation but the

interval of time between these events varies greatly for individual

nuclei; neurons of the lateral nucleus, for example, arise early but

the nuclear area is not recognizable until just before birth.

In the present study sequence in origin of preoptic neurons

generally parallels their appearance as nuclear areas as reported by

Niimi et al. (1962); in general, the nuclear area is recognizable at

the end of the period of origin of its constituent neurons. The medial

preoptic nucleus, however, is recognizable precociously, before all of

its neuron precursors have undergone final division. Such a brief

interval, or in fact overlap, between proliferation and nuclear recog­

nition has been reported where neuroblasts migrate only short distances

(medial habenula, Angevine 1970a). The lateromedial gradient in

preoptic nuclear origin (lateral, E10-16; medial, E10-18) is paralleled

by a similar gradient in recognition, the lateral nucleus appearing on

E15 and the medial early on El6. The mediolateral gradient of origin 108 when periventricular nucleus is compared to medial, however, does not

correlate with any differentiation gradient; the suprachiasmatic and

medial nuclei are recognizable simultaneously (early El6) and the

periventricular nucleus appears slightly later (El6). Thus, in the preoptic area, as in other areas, there is a general but not exact correspondence between sequence in time of origin and time of nuclear

recognition.

Time and site of origin of neurons can be determined autoradio- graphically but in .order to establish the precise temporal relationship of proliferation to differentiation, correlated electron microscopic and Golgi techniques must be employed. Such a study of development of stellate and basket cells of the cerebellar cortex of the rhesus monkey by Rakic (1973) has shown that the interval between proliferation and differentiation may be a function not only of neuronal type but also of adult position. The last neurons to arise lie most superficially and also have the longest proliferation-differentiation interval. But if a few neurons come to lie deeper than most others formed on the same day, they differentiate sooner, suggesting that it is not an intrinsic property of the neuroblast which determines the time of differentiation but the milieu into which the young neuron comes to lie. When the milieu is appropriate, differentiation is apparently triggered. Simil­ arly, Morest (1969b) has noted that dendritic differentiation of a particular neuron usually coincides with growth into the area of the axonal plexus which will eventually stimulate that neuron. That these 109

afferent endings may instigate dendritic differentiation by physico- chemical mechanisms or by providing a mechanical framework for the growing processes is an intriguing idea now under investigation in many laboratories, especially where tissue culture studies are carried out

(LaVail and Wolf, in press).

Nucleus Accumbens - Septal or Striatal?

There is some question whether the nucleus accumbens is a septal nucleus. Although it extends into the septal region, many investigators consider it part of the striatal complex (Young 1936, Fox 19^0, Lauer

19^5? Nauta and Haymaker 1969), since it is continuous with the head of the caudate and contains similar neurons. Phylogenetically it is apparently unlike the rest of the septal area, as it is relatively large in the human brain compared to the other septal nuclei which show little elaboration vis-a-vis their counterparts in other mammals.

It is of interest therefore to compare the time of origin of the nucleus accumbens to that of the septal area on the one hand and the striatal complex on the other. As reported elsewhere in this paper, nucleus accumbens has a much longer period of proliferation (E10-P4) than the other septal nuclei, which have end points seldom beyond El6.

Although neurons of the caudate and putamen were previously reported to arise early (E12-15, Sidman and Angevine 1962) the end-point of proliferation was not determined. Examination of the animal injected on PI in this study revealed that several striatal neurons, especially in the head of the caudate, were heavily labelled, similar to those of the nucleus accumbens. The late forming neurons of the basal ganglia 110

are derived from subventricular cells of the ganglionic eminence, at

least in the human. Probably, in the mouse, accumbens and caudate

neurons are similarly formed, but this fact could not be established.

Similar periods of labelling in accumbens and caudate were

reported earlier in the postnatal rat (Das and Altman 1970), where the

site of origin is the subventricular zone on the lateral surface of

lateral ventricle. Similarities in time of origin alone, however, are

insufficient to conclusively classify a nucleus as belonging to a

particular region, since nuclei within one region often have very

different times of origin (see Telodiencephalic Boundary). All that

can be concluded is that the nucleus accumbens appears more related

to the striatal complex than to the septal area on the basis of time

of neuron origin.

Phylogenetic Parallels of Time of Origin

Since developing structures often, but not necessarily, pass through stages similar to those which the species has presumably under­

gone during its phylogenetic development, the oldest neural structures

(the ones most differentiated in simple species) might be expected to

have the earliest times of origin. The cerebral cortex offers an outstanding example; neurons of the outer layers are last to arise and

are phylogenetically most recent (Angevine and Sidman 1961, Berry and

Eayrs 1963? Berry and Rogers 1965). Similarly, in the hippocampus, the prolonged proliferation of granule cells may be an ontogenetic expres­ sion of the conspicuous increase in the size and neuronal population of this layer in placental compared to nonplacental mammals (Angevine 1965). The early origin and subsequent differentiation of the ventral thalamus finds a parallel in the precocity of this diencephalic compo­ nent in vertebrate suprasegmental sensorimotor controls (Angevine 1970a).

Also, in the olfactory formation neurons of the phylogenetically older accessory bulb arise prior to their counterparts in the newer olfactory bulb (Hinds 1968a). Although these few examples seem to support the ontogenetic recapitulation of phylogenetic events, there are many exceptions.

The present finding that many basal forebrain neurons arise early accords with their early appearance phylogenetically. A compar­ ison of time of origin of the septal nuclei with their apparent phylogenetic trend in primates (Andy and Stephan 1966) shows a rough parallel. The septal nuclei and their times of origin in the mouse listed in order of increasing phylogenetic trend (size relative to entire septal area) are: triangular (E10-15), nucleus of the anterior commissure (E10-16), medial (ElO-lU), fimbrial (E10-15), nucleus of the diagonal band (E10-16), lateral (E10-18), and nucleus of the stria terminalis (E10-17). In general those nuclei which have decreased in size or remained stable (triangular and anterior commissure) have early periods of origin and those which have increased have later and longer times of origin. The nucleus accumbens, not analyzed by Andy and Stephan offers a very tempting example of such phylogenetic/ ontogenetic parallelism (see Nucleus Accumbens - Septal or Striatal?). 112

The Question of the Telodiencephalic Boundary

Most neuroerabryologists recognize the usefulness of a generalized classification of brain development into a two-vesicle (archencephalon

and deuterencephalon; acrencephalon and chordencephalon), a three-vesicle

(prosencephalon, mesencephalon, and rhombencephalon) and finally a five-

vesicle stage (telencephalon, diencephalon, mesencephalon, metencephalon,

and myelencephalon). Others consider the more numerous neuromeres, bulges in the neural wall first described by Karl Ernst von Baer (1828), to be the fundamental divisions (Bergquist and Kallen 195^+). In either case, the boundaries of these vesicles or neuromeres are not arbitrary; they appear in fairly regular positions during CNS development of most vertebrates. Comparative anatomists, in an attempt to establish homologies, have used these embryonic structures as landmarks to describe adult nervous systems. Since embryonic folds and bulges are difficult to follow throughout development, many disagreements as to the exact correspondence of certain structures and a cumbersome terminology have developed (see Results - Septal Area).

One of these currently unresolved issues is the position of the telodiencephalic boundary, especially in the greatly elaborated mammalian brain. Wilhelm His (1888) arbitrarily drew the boundary such that most of the hypothalamus (except the mammillary bodies) is telencephalic (see

Preoptic Area). Although most anatomists disagree with the boundary of

His, an alternative acceptable to everyone has not been found. Those who consider the longitudinal subdivisions of the diencephalon more important in establishing homologies (Kuhlenbeck and Haymaker 19^9) draw the line 113

from the transverse velum to the commissural plate (anterior commissure

in adult) making the preoptic area diencephalic. In contrast, those who

use neuromeres to establish homologies (Rudebeck 19^+5) draw the line to

the preoptic recess, making the preoptic area telencephalic. Similarly

von Kupffer (1906) defined the boundary in relation to the anterior

intraencephalic sulcus, which in the adult corresponds to a line drawn

from the foramen of Monro to the optic recess; this boundary, presented

in many current textbooks (Truex and Carpenter 1969, Crosby, Humphrey,

and Lauer 1962), also makes the preoptic area telencephalic.

Whether the proptic area is better considered telencephalic or

diencephalic could be determined if some characteristic of the preoptic

area were more like the telencephalon on the one hand or the dienceph-

alon on the other. That time of origin may be one such characteristic will be considered. Results of the present study show that preoptic neurons arise from E10-18, but mostly from E12-1U. This period of origin is similar to that of the telencephalic nuclei which border it, being slightly later than septal nuclei (except nucleus accumbens, which see), slightly earlier than anterior olfactory nucleus, and about the same as the olfactory tubercle. An "outside-in" gradient is also similar to that in the septal area. Thus, in the mouse, there is no basis for separating the preoptic area from the telencephalon according to its time of origin.

Although the remainder of the hypothalamus was not analyzed in this study, a comparison with preliminary reports by Angevine (1970c) suggests that the preoptic area arises later than the adjacent supraoptic area (E12-l^ compared to E10-11) but concurrently with 114 caudal levels of hypothalamus. Even though time of preoptic origin is not consistent with the rostrocaudal gradient noted in the hypothalamus, it has similar lateromedial and dorsoventral gradients. Differences in time of origin at preoptic vs. supraoptic levels would suggest that the preoptic is more closely related to the telencephalon, but studies in the rat (ifft 1972) show times of origin similar at all levels of hypothalamus, including preoptic area. In studies of diencephalic development in the hamster, Keyser (1972) concludes that although lateromedial and dorsoventral gradients are evident, there is no overall rostrocaudal gradient in time of origin in the hypothalamus; rather, each level has a characteristic time of origin. Examination of his figure (p. 12^+), however, reveals that although labelling at the preoptic level is dissimilar to that at the adjacent supraoptic level it is similar to that at the mammillary level, as in the mouse (see above).

Thus, time of origin of preoptic neurons is similar to that of adjacent telencephalic neurons in the mouse, but similarities also exist in the other direction, with diencephalon. Preoptic origin resembles that of adjacent diencepahlic neurons (supraoptic hypothalamus) in the rat, but is longer and later than that of supraoptic neurons in the hamster and mouse. A finding in all studies is that the hypothalamus, including the preoptic area, arises in a dorsoventral and lateromedial

("outside-in") sequence. Therefore it is still not possible, on the basis of time of origin alone, to conclude whether preoptic neurons are telencephalic or diencephalic. 115

The assumption underlying the above discussion is that some fundamental difference between telencephalon and diencephalon will be reflected in their times of origin. Now that extensive time-tables of proliferation of telencephalic nuclei (see Table 3) and diencephalic nuclei (Angevine 1970a, 1970c) are available it is apparent that, although gradients within large areas can be detected, individual nuclei have specific times of origin, as distinctive as those of components of the hippocampal region (Angevine 1965), for example. Therefore, there is as much variation in time of origin of nuclei within the telen­ cephalon or diencephalon as between them. In fact, within the telencephalon, cortical and nuclear areas have fundamentally different patterns of proliferation, "inside-out" vs. "outside-in", which represent transposition of cell nuclei on the one hand arid simple displacement of cell nuclei on the other. Within the diencephalon, even opposite gradients, caudorostral (thalamus) vs. rostrocaudal

(hypothalamus), ventrodorsal (ventral thalamus, dorsal thalamus, and epithalamus) vs. dorsoventral (hypothalamus) coexist!

Even though no gradients or periods of origin distinguish the telencephalon from the diencephalon, if there are differences based on other criteria, such as types of neurons, fiber connections, or func­ tional significance, then a boundary or concept of a boundary zone is not arbitrary. Descriptions of diencephalic or basal telencephalic nuclei, however, stress again and again the continuities between these areas and emphasize that some telencephalic nuclei are inseparable from diencephalic ones on the basis of architectonics or fiber connections. 116

The septal, preoptic, and hypothalamic areas are said to form a func­

tional continuum. Since no differences on any of these cardinal

neuroanatomical criteria exist, unless other characteristics can clearly

distinguish the diencephalon and telencephalon, the subdivision of the

adult forebrain based solely on folds and bulges in the embryo is super­

fluous .

Some would argue that even if there is no fundamental difference

in diencephalon and telencephalon in the adult these boundaries are

essential to the comparative anatomist who hopes to establish homologies

between brains throughout the phyletic series. In a study of neuromery,

heterochrony (differences in development of matrix layer), and other

ways of demarcating embryonic regions, Keyser (1972) concludes that "The

heterogeneity of factors leading to the development of the diencephalic

sulci hardly leaves any doubt as to the unreliability of grooves in the

ventricular relief as a criterion of subdivision with the diencephalon.. ."

(p. 6l). Furthermore, the variability and difficulty in following such

grooves throughout development make this a poor basis for establishing

homologies and lead to disagreement among comparative neuroanatomists

(see above).

That there are only slight differences in telencephalic vs. diencephalic nuclei on the basis of cytoarchitectonics, fiber connec­ tions, functional significance, or developmental events, such as time of origin, supports the view of Bergquist and Kallen (195*0 that "If a tel-diencephalic limit is fixed -- e.g., following Johnston's ('09) method -- it will be an artificial one and no morphological importance can be ascribed to it," (p. 6U9). APPENDIX

AUTORADIOGRAPHIC RECONSTRUCTIONS AND GRAPHS

The following autoradiographic reconstructions and graphs indicate the population of neurons whose precursors divided for the final time on the day of H -thymidine injection, from the tenth day of gestation to the fourth postnatal day. The day of injection, the age of the animal (in months) when sacrificed, and the transverse level from which the section was taken are indicated by the numbers appearing in the upper right hand corner of each illustration.

E10-P6-V, for example, is the fifth level of the brain of an animal 3 which received H -thymidine on the tenth day of embryogenesis and was killed at six months of age. Figs. 2-U6 illustrate the exact position of cell nuclei of heavily (solid circles) and lightly (open circles) labelled neurons in the anterior olfactory nucleus, olfactory tubercle, islands of Calleja, preoptic and septal areas, and nucleus of the lateral olfactory tract. Figs. U7-5& are graphic representations of the number of heavily labelled neurons on each day of development for each nuclear area. Table 3 lists all nuclei and tracts whose abbreviations appear on the autoradiograms and Fig. 1 illustrates the position of the five transverse levels (i-V) studied.

117 118

TABLE 3

LIST OF ABBREVIATIONS

AOd Nuc. olfactorius anterior, pars dorsalis AOe Nuc. olfactorius anterior, pars externa AOl Nuc. olfactorius anterior, pars lateralis AOm Nuc. olfactorius anterior, pars medialis AOp Nuc. olfactorius anterior, pars posterior 0A Commissura anterior CC Corpus callosum CI Capsula interna CO Chiasma opticum CP Nuc. caudatus putamen IC Insulae Callejae ICm Insula magna Callejae Lpl Lamina plexiformis, tuberculum olfactorium Lpo Lamina polymorphica, tuberculum olfactorium Lpy Lamina pyramidalis, tuberculum olfactorium NCA Nuc. commissurae anterioris NST Nuc. interstitialis striae terminalis NTD Nuc. tractus diagonalis Brocae NTOL Nuc. tractus olfactorius lateralis PI Nuc. preopticus lateralis Pm Nuc. preopticus medialis Pmg Nuc. preopticus magnocellularis Pmn Nuc. preopticus medianus Pp Nuc. preopticus periventricularis Ps Nuc. preopticus suprachiasmatica Sa Nuc. septalis accumbens Sf Nuc. septalis fimbrialis SI Nuc. septalis lateralis Sm Nuc. septalis medialis St Nuc. septalis triangularis TO Tuberculum olfactorium TOI Tractus olfactorius intermedius TOL Tractus olfactorius lateralis 119

Fig. 1 Basal (above) and sagittal (below) views of the mouse brain illustrating levels I-V used to prepare autoradiographic reconstructions. (Drawings modified from the Atlas of the Mouse Brain and Spinal Cord by Sidman, Angevine and Taber Pierce 1971, with the permission of the authors and publisher). 120

Ell "PI "I

/ o AOd

TOI

AOm? AOp

AOI

A0«

EI2-P6-I

AOd

TOI .

AOm, AOt

AOp

AOe

in anterior olfactory nucleus on Ell Fig. 2 Pattern of labelling (above) and early on E12 (below). 121

E12-PI-I

Ctf 8

AOm '§3 A r , I J? o o -f Jo°. i "

EI3-PI-I

Oo'o AOd

AOm TOI

o y±\ fc« AOI I.

,\°0 AOe

Fig. 3 Pattern of labelling in anterior olfactory nucleus late on E12 (above) and on E13 (below). 122

EI4-P6-I

AOd

AOfTV " /£

TOI

AOI

AOe

EI4-PI-I

• AOd

TOI AOm

AOI

AOp

AOe

Fig. k Pattern of labelling in anterior olfactory nucleus early on ElU (above) and late on ElU (below). 123

EI5-P9-I

AOd

TOI AOm AOp

AOI

AOt

EI6-P3-I

AOd

TOI AOm AOp

AOI

AOe

Fig. 5 Pattern of labelling in anterior olfactory nucleus on E15 (above) and early on El6 (below). 12k

EI7-P3-I

AOd

/ \

AOm I TOI AOp

AO!

AO*

EI8-PI-I

AOd

TOI

AOm AOp

AOI

AO«

Fig. 6 Pattern of labelling in anterior olfactory nucleus on E17 (above) and early on El8 (below). 125

EI8-P3-I

AOd

AOm AOp TOI

AOI

AOs

PI-P3-I

AOd

AOm

TOI AOp

AOI

AO*

Fig. 7 Pattern of labelling in anterior olfactory nucleus late on EL8 (above) and on postnatal day one (below). 126

CC

EI0-P7-II

CA

NTD So

L po

TO

Fig. 8 Pattern of labelling at rostral septal level on ElO. 127

cc

Sm

CA

So

NTD

Fig. 9 Pattern of labelling at rostral septal level on Ell. 128

Fig. 10 Pattern of labelling at rostral septal level early on E12. 129

o

EI2-PI-II

CA

NTD

So

Fig. 11 Pattern of labelling at rostral septal level late on E12. 130

cc

Sm

CA

Sa NTD

Lpo

Fig. 12 Pattern of labelling at rostral septal level on E13. 131

ec

EI4-P6-II

CA

NTD

riC *,Lpo -

TO"

Fig. 13 Pattern of labelling at rostral septal level early on El4. 132

cc

EI4-PI -II

Sm *% •.

CA

So NTO

Lpo

Fig. lU Pattern of labelling at rostral septal level late on El^t-. 133

Fig. 15 Pattern of labelling at rostral septal level on E15. 134

CC

EI6-P2-II

Sm

NTD

CA

L po

TO'

Fig. 16 Pattern of labelling at rostral septal level early on El6. 135

CC

Sm

CA r> 'Cm.

NTO So

IC—,

Lpo

ilpl TO-

Fig. 17 Pattern of labelling at rostral septal level late on El6. 136

CQ

EI7-P3-II

SmI

NTD

Lpo j»r 17. TO'

Fig. 18 Pattern of labelling at rostral septal level on E17. 137

EI8-PI-II

Sm

NTD

So

Lpo

Fig. 19 Pattern of labelling at rostral septal level early on El8. 138

CC

EI8-P3-II

Sm

NTO

Lpo

TO'

Fig. 20 Pattern of labelling at rostral septal level late on El8. 139

l.-,T|Cm

Fig. 21 Pattern of labelling at rostral septal level on postnatal day one. Ibo

Fig. 22 Pattern of labelling at rostral septal level on postnatal day four. lUl

cc

EI0-P7-III

Sm I

cr NST

Pmn* /

Pmg NTO

TCr

Fig. 23 Pattern of labelling at middle septal and rostral preoptic level on E10. 142

cc

EII-PI-III 9m

NST

CA

o o •/ Pmn'

NST CA GP

Pm

NTD Pmg

CO

Fig. 2b Pattern of labelling at middle septal and rostral preoptic level on Ell. 1^3

Fig. 25 Pattern of labelling at middle septal and rostral preoptic level early on E12. 144

Fig. 26 Pattern of labelling at middle septal and rostral preoptic level late on E12. l!+5

EI3-PI-III

. . 8

o°o -

NST

CA

NST CA Pmn'

%''-CP

Pm

NTD

. _ '• • \

Fig. 27 Pattern of labelling at middle septal and rostral preoptic level on E13• lU6

CC

EI4-P6-III

SmI

NST

CA

Pmn NST

Pm

NTD

Pmg

'Ps

CO

Fig. 28 Pattern of labelling at middle septal and rostral preoptic level early on El^. 147

ec

EI4-PI-III

t/o'

NST

CA

\ •/ Pmn. NST

Pm

NTD Pmg L* P»

CO / v

c Lpoc Lpl

TO

Fig. 29 Pattern of labelling at middle septal and rostral preoptic level late on El4. 148

Fig. 30 Pattern of labelling at middle septal and rostral preoptic level on E15- 149

cc

EI6-P2-III

NST

NST Pmn

Pm

NTD

Pmg

CO

LRo

Fig. 31 Pattern of labelling at middle septal and rostral preoptic level early on El6. 150

EI6-P3-III

Sffl

NST

NST Pmn

Pfn

NTD Pmg

CO

Fig. 32 Pattern of labelling at middle septal and rostral preoptic level late on El6. 151

CC

EI7-P3-III

Sm

NST

CA

Pitin

NST

Pm

NTO Pmg

CO.

TO"

Fig. 33 Pattern of labelling at middle septal and rostral preoptic level on E17. 152

•y i^NCA\

Fig. 3^ Pattern of labelling at caudal septal and preoptic level on E10. 153

CC

Ell-Pl-IV

NST ,NC$

Pm

NTD

CO

Fig. 35 Pattern of labelling at caudal septal and preoptic level on Ell. 154

EI2-P6-IV

O o °"0°oVSt _ «« °O 0 \ &l£-n i °o \>„° • • I "°S\& K °„ j»V Vo** ov^ 0*11

Nsr oo

Fig. 36 Pattern of labelling at caudal septal and preoptic level early on E12. 155

Fig. 37 Pattern of labelling at caudal septal and preoptic level late on E12. 156

CC

EI3-PI-IV 0 ^

Sf

NST

CA

NCA

Pmg

?/ P» NTD

CO

Fig. 38 Pattern of labelling at caudal septal and preoptic level on E13. 157

CC

EI4-P6-IV

NST NCA.,

Pmn i

Pm

Pmg

NTD

CO

Fig. 39 Pattern of labelling at caudal septal and preoptic level early on El^. 158

NCA NST

CA

Pmn

Pm

NTO

CO

Hg' k° fa««n of labelii late on Bib. lng at oau3al septal and preoptic level 159

CC

EI5-P9-IV

NST NCA

Pmn

Pm

Pmg NTD

CO

Fig. ^1 Pattern of labelling at caudal septal and preoptic level on E15- EI6-P2-IV

J-

iCA NCA >

.Pmi

NST

Pm

NTD Pmg

CO

Fig. b2 Pattern of labelling at caudal septal and preoptic level early on El6. l6l

EI6-P3-IV

NCA

NST

\

Pm

Pmg NTO

Ps

CO

Fig. ^3 Pattern of labelling at caudal septal and preoptic level late on El6. 162

CQ

EI7-P3-IV

NST NCA

IPnw

|PPI Pm

Pmg

NTD

Fig. M* Pattern of labelling at caudal septal and preoptic level on E17- 163

cc

EI8-PI-IV

NCA

NST CA

Pmn

Pm

Ping

NTD

CO

Fig* ^5 Pattern of labelling at caudal septal and preoptic level early on El8. 164

NTOL

NTOL

.NTOL

NTOL

Pattern of labelling in the nucleus of the lateral olfactory tract on (from upper right to lower left) E10, Ell, early E12, E13, early El^, and late El4. 165

40" PARS EXTERNA NUC. OLFACTORIUS ANTERIOR 30* CO z °or 20" z> UJ Z 10 LeveI I Q UJ _J ElO

140" PARS LATERALIS

>120" NUC. OLFACTORIUS ANTERIOR

X |oo" li-

80"

60"

40"

Level I

ElO DAY OF DEVELOPMENT

Fig. hi Number of heavily labelled neurons on each day of development in the anterior olfactory nucleus, pars externa (above) and pars lateralis (below). 166

30" PARS MEDIALIS

NUC. OLFACTORIUS ANTERIOR 20"

O 10" Level I

UJ 60" PARS DORSALIS NUC. OLFACTORIUS ANTERIOR

30"

Level I

ElO

60" PARS POSTERIOR 45' NUC. OLFACTORIUS ANTERIOR

30"

Level I

ElO DAY OF DEVELOPMENT

Fig. 48 Kumber of heavily labelled neurons on each day of development in the anterior olfactory nucleus, pars medialis (above), pars dorsalis (middle), and pars posterior (below). 167

45"

NUC. TRACTUS OLFACTORII LATERALIS

£C 15" Level V

LU —I W 45" CD LAMINA POLYMORPHICA < TUBERCULUM OLFACTORIUM -1 3CT >- Level II > < LU X li. o

60" LAMINA PYRAMIDALIS TUBERCULUM OLFACTORIUM

30"

Level II

DAY OF DEVELOPMENT

Fig. i+9 Number of heavily labelled neurons on each day of development in nucleus of the lateral olfactory tract (above), and olfactory tubercle, polymorph layer (middle) and pyramidal layer (below). 168

INSULAE CALLEJAE

Level II

INSULA MAGNA CALLEJAE _| 45

Level II _l

X 100" Lu NUC. COMMISSURAE ANTERIORIS

40'

26 Level IV

ElO II 12 13 14 15 16 17 18 PI 2 3 4 5 DAY OF DEVELOPMENT

Fig. 50 Number of heavily labelled neurons on each day of development in islands of Calleja (above), insula magna (middle), and nucleus of the anterior commissure (below). 169

15 NUC. PREOPTICUS MEDIANUS 1(51 / /\\ / 7 M Level III \\ Level IV

NUC. PREOPTICUS PERIVENTRICULAR IS

Level in Level IV

CC EIO

NUC. PREOPTICUS SUPRACHIASMATICA

Level III Level IV

12 13 14 15 16 17 DAY OF DEVELOPMENT

Fig. 51 Number of heavily labelled neurons on each day of development in median (above), periventricular (middle), and suprachiasmatic (below) preoptic nuclei. 170

100"

NUC. PREOPTICUS LATERALIS ecr

60"

UJ 40

Level III — —Level IV

ElO 160

>- 140" NUC. PREOPTICUS MEDIALIS

<120"

100

80 LU 60"

40

—Level in 20 — Level IV

ElO II 12 13 14 15 16 17 18 PI DAY OF DEVELOPMENT

Fig. 52 Number of heavily labelled neurons on each day of development in lateral (above) and medial (below) preoptic nuclei. 171

40'

NUC. PREOPTICUS MAGNOCELLULARIS

Level III —Level IV

ElO LU

—1 105"

90" NUC. TRACTUS DIAGONALIS 8R0CAE

U. O 60"

m 45

Z 30' Level II Level III Level IV

ElO II 12 13 14 15 16 17 18 DAY OF DEVELOPMENT

Fig. 53 Number of heavily labelled neurons on each day of development in magnocellular preoptic nucleus (above) and nucleus of the diagonal band of Broca (below). 172

30 NUC. SEPTALIS MEDIALIS 20"

Level II

ElO

300"

UJ -J 270 NUC. SEPTALIS LATERALIS

240"

>- 210" —I

< 180"

U. 150"

CC 120 UJ

90*

60 Level II Level III 30" — Level IV

ElO DAY OF DEVELOPMENT

Fig. 5*+ Number of heavily labelled neurons on each day of development in medial (above) and lateral septal nuclei (below). 173

100"

80" NUC. SEPTALIS TRIANGULARIS CO z 60

P 40"

Level III 20" Level IV

LU ElO II 12 13 14 15 16 IT 18 PI

140

120" NUC. SEPTALIS FIMBRIALIS

X 100

80

QQ 60"

Z 40"

Level III 20" Level IV

ElO DAY OF DEVELOPMENT

Fig. 55 Number of heavily labelled neurons on each day of development in triangular (above) and fimbrial (below) septal nuclei. Yjk

160"

NUC. INTERSTITIALIS STRIAE TERMINALIS 120"

Z 80"

_J 60

LU m 40

_l Level III >- 2°" —Level IV _J

ElO LU

U.IOO" NUC. SEPTALIS ACCUMBENS a: eo"

i 60"

40"

20' Level II

DAY OF DEVELOPMENT

Fig. 56 Pattern of heavily labelled neurons on each day of development in interstitial nucleus of the stria terminalis (above) and nucleus accumbens septi (below). LITERATURE CITED

Allen, Ezra 1912 The cessation of mitosis in the central nervous system of the albino rat. J. Comp. Neur., 22: 5^7-568.

Allison, A. C. 1953 The structure of the olfactory bulb and its relationship to the olfactory pathways in the rabbit and rat. J. Comp. Neur., 98: 309-353.

Altman, Joseph 1966 Autoradiographic and histological studies of postnatal neurogenesis. II. A longitudinal investigation of the kinetics, migration, and transformation of cells incorporating tritiated thymidine in infant rats, with special reference to postnatal neurogenesis in some brain regions. J. Comp. Neur., 128: U31-V73-

1967 Postnatal growth and differentiation of the mammalian brain, with implications for a morphological theory of memory. In: The Neurosciences: A Study Program, G. C. Quarton, T. Melnechuk, and F. 0. Schmitt (eds.). Rockefeller Univ. Press, New York, pp. 723-7^3.

1969a Autoradiographic and histological studies of postnatal neurogenesis. III. Dating and time of production and onset of differentiation of cerebellar microneurons in rats. J. Comp. Neur., 136: 269-293.

1969b Autoradiographic and histological studies of postnatal neurogenesis. IV. Cell proliferation and migration in the anterior forebrain, with special reference to persisting neurogenesis in the olfactory bulb. J. Comp. Neur., 137: ^33-^57.

' 1970 Postnatal neurogenesis and the problem of neural plasti­ city. In: Developmental Neurobiology. W. A. Himwich, and D. Purpura (eds.). Charles C Thomas, Springfield, pp. 197-237.

1972a Postnatal development of the cerebellar cortex in the rat. I. The external germinal layer and the transitional molecular layer. J. Comp. Neur., 1^5: 353-397.

175 176

Altman, Joseph 1972b Postnatal development of the cerebellar cortex in the rat. II. Phases in the maturation of Purkinje cells and of the molecular layer. J. Comp. Neur., 1^5: 399~^6U.

1972c Postnatal development of the cerebellar cortex in the rat. III. Maturation of the components of the granular layer. J. Comp. Neur., lU5: h6^-3lh.

Altman, Joseph and Gopal D. Das I96U Autoradiographic examination of the effects of enriched environment on the rate of glial multi­ plication in the adult rat brain. Nature, 20k: Il6l-ll63.

1965a Autoradiographic and histological evidence of postnatal hippocampal neurogenesis in rats. J. Comp. Neur., 12^: 319-336.

1965b Post-natal origin of microneurons in the rat brain. Nature, 207: 953-956.

1966 Autoradiographic and histological studies of postnatal neurogenesis. I. A longitudinal investigation of the kinetics, migration, and transformation of cells incorporating tritiated thymidine in neonate rats, with special reference to postnatal neurogenesis in some brain regions. J. Comp. Neur., 126: 337-389.

1967 Post-natal neurogenesis in the guinea pig. Nature, 21k: IO98-IIOI.

Altman, Joseph, Gopal D. Das, and William J. Anderson 1968 Effects of infantile handling on morphological development of the rat brain: An exploratory study. Devel. Psychobiol., 1: 10-20.

Altmann, Richard l88l Vorlaufige Mittheilung uber embryonales Wachs- thum. Flieg. Blatt., Leipzig (as cited in Hinds 1967).

Andy, Orlando J. and Heinz Stephan 1966 Septal nuclei in primate phylogeny. A quantitative investigation. J. Comp. Neur., 126: 157-171.

Angevine, Jay B., Jr. 1963 Autoradiographic study of histogenesis in the hippocampal formation of the mouse. Anat. Rec., 1^5: 201. 177

Angevi'ne, Jay B., Jr. 196U Autoradiographic study of histogenesis in the area dentata of the cerebral cortex in the mouse. Anat. Rec., 1U8: 255.

1965 Time of neuron origin in the hippocampal region. An autoradiographic study in the mouse. Exp. Neur., Suppl. 2 1-70.

1970a Time of neuron origin in the diencephalon of the mouse. An autoradiographic study. J. Comp. Neur., 139: 129-187.

1970b Critical cellular events in the shaping of neural centers. In: The Neurosciences, Second Study Session. G. C. Quarton et al. (eds.). The Rockefeller University Press, New York, pp. 62-73.

1970c Gradients in time of neuron origin in mouse diencephalon. Anat. Rec., 166: 272.

Angevine, Jay B., Jr. and Richard L. Sidman 1961 Autoradiographic study of cell migration during histogenesis of cerebral cortex in the mouse. Nature, 192: 766-768.

1962 Autoradiographic study of histogenesis in the cerebral cortex of the mouse. Anat. Rec., 1^2: 210.

Atlas, M. and V. P. Bond 1965 The cell generation cycle of the eleven-day mouse embryo. J. Cell Biol, 26: 19-2U.

Baer, Karl Ernst von 1828 Ueber die Entwicklungsgeschichte der Thiere Bd 1-3, Konigsberg.

Bailey, Percival and Harvey Cushing 1926 A classification of the tumors of the glioma group on a histogenetic basis with a corre­ lated study of prognosis. Lippincott, Philadelphia, 175 pp.

Ban, Tadayasu and Katuyo Zyo 1962 Experimental studies on the fiber connections of the rhinencephalon. I. Albino rat. Med. J. Asaka Univ., 12: 385-^2^.

Baserga, Renato and Daniel Malamud 19&9 Modern methods in experimental pathology. Autoradiography, techniques and application. Harper and Row, New York, 28l pp. 178

Becquerel, Henri 1896 Sur les radiations emises par phosphorescence. Compt. Rend., 122: k20-k21.

Belanger, Leonard F. and Charles P. LeBlond 19U6 A method for locating radioactive elements in tissues by covering histological sections with a photographic emulsion. Endocrinology, 39- 8-13.

Bennett, Edward L., Marian C. Diamond, David Krech, and Mark R. Rosenzweig 196^ Chemical and anatomical plasticity of brain. Science, lU6: 610-619.

Bergquist, Harry and Bengt Kallen 195^- Notes on the early histogenesis and morphogenesis of the central nervous system in vertebrates. J. Comp. Neur., 100: 627-660.

Berry, M. and J. T. Eayrs 1963 Histogenesis of the cerebral cortex. Nature, 197: 984-985.

Berry, M. and Andrew ¥. Rogers 1965 The migration of neuroblasts in the developing cerebral cortex. J. Anat., 99: 691-709 (Lond.).

Berry, M., Andrew W. Rogers, and J. T. Eayrs 1963 The pattern and mechanism of migration of the neuroblasts of the developing cerebral cortex. J. Anat., 98: 291-292.

_ 196k Pattern of cell migration during cortical histogenesis. Nature, 203: 591-593.

Birmingham Revision of the BNA, (BR) Anatomical Society of Great Britain and Ireland 1933 Final report of the committee appointed by the Anatomical Society of Great Britain and Ireland on June 22, 1928. Robert Maclehose and Co. Ltd., University Press, Glasgow.

BNA of 1895 His, W. 1895 Die anatomische Nomenclatur. Nomina anat- omica. Separat-Abzug aus Archiv fur Anatomie und Physiologie. Anatomische Abteilung. Supplement-Band.

Boulder Committee 1970 Embryonic vertebrate central nervous system: Revised terminology. Anat. Rec., 166: 257-262.

Broca, Pierre Paul 1878 Anatomie comparee des circonvolutions ce're- brales. Le grand lobe limbique et la scissure limbique dans le serie des mammiferes. Rev. Anthropol., Ser. 2, I: 385-^98. 179

Broca, Pierre Paul 1879 Recherches sur les centres olfactifs, Rev. d'Anthrop., 2: 385-^55- (Reprinted in Memoires d'Anthropologic de Paul Broca, vol. 5. C. Reinwald, Paris, 1888, 8U0 pp.)

Brodal, Alf 19^8 The origin of the fibers of the anterior commissure in the rat. Experimental studies. J. Comp. Neur., 88: 157-205.

Brun, Arne 1965 The subpial granular layer of the foetal cerebral cortex in man. Its ontogeny and significance in congenital cortical malformations. Acta. Path, et Microbiol-., 179: 1-98.

Caro, Lucien G. 1962 High-resolution autoradiography. II. The problem of resolution. J. Cell Biol., 15: 189-199-

Christ, J. F. 1969 Derivation and boundaries of the hypothalamus, with atlas of hypothalamic grisea. In: The Hypothalamus, W. Haymaker, E. Anderson, W. J. H. Nauta (eds.), Charles C Thomas, Springfield, pp. 13-60.

Coggeshall, Richard E. 1964 A study of diencephalic development in the albino rat. J. Comp. Neur., 122: 2UI-269.

Coleman, E. J. 1965 A simplified autoradiographic dipping procedure - slides handled in groups of five. Stain Technol., 40: 2b0-2kl.

Cowan, W. Maxwell, Alexander H. Martin, and Eleanor Wenger 1968 Mitotic patterns in the optic tectum of the chick during normal development and after removal of the optic vesicle. J. Exp. Zool., 169: 71-92.

Creps, Elaine S. and Jay B. Angevine, Jr. 1971 Time of origin of basal forebrain neurons in the mouse. Anat. Rec., 169: 302.

Crosby, Elizabeth C. 1917 The forebrain of Alligator mississippiensis. J. Comp. Neur. 27: 325-^-02.

Crosby, Elizabeth C. and Tryphena Humphrey 19^-1 Studies of the vertebrate telencephalon. II. The nuclear pattern of the anterior olfactory nucleus, tuberculum olfactorium, and the amygdaloid complex in adult man. J. Comp. Neur., 7^: 309-352.

Crosby, Elizabeth, Tryphena Humphrey, and Edward W. Lauer 1962 Correlative Anatomy of the Nervous System. The Macmillan Co. New York, 696 pp. l8o

Das, Gopal D. and Joseph Altman 1970 Postnatal neurogenesis in the caudate nucleus and nucleus accumbens septi in the rat. Brain Res., 21: 122-127.

DeLong, Robert G. and Richard L. Sidman 1962 Effects of eye removal at birth on histogenesis of the mouse superior colliculus: An autoradiographic analysis with tritiated thymidine. J. Comp. Neur., 118: 205-223.

Diamond, Marian C., David Krech, and Mark R. Rosenzweig 1964 The effects of an enriched environment on the histology of the rat cerebral cortex. J. Comp. Neur., 123: 111-120.

Diamond, Marian C., F. Law, H. Rhodes, B. Lindner, Mark R. Rosenzweig, David Krech, and L. Bennett 1966 Increases in cortical depth and glia numbers in rats subjected to enriched environment. J. Comp. Neur., 128: 117-125.

Edinger, Ludwig 1908-1911 Vorlesungen uber den Bau der nervosen Centralorgane des Menschen und der Thiere. Vogel, Leipzig.

Ellenberger, Carl, Joseph Hanaway, Jr., and Martin G. Netsky 19&9 Embryogenesis of the inferior olivary nucleus in the rat: A radioautographic study and a re-evaluation of the rhombic lip. J. Comp. Neur., 137: 71-87.

Elliot Smith, Grafton 1896 Morphology of the true "limbic lobe" corpus callosum, septum pelucidum and fornix. J. Anat., 30: 185-205.

1903 Zuckerkandl on the phylogeny of the corpus callosum. Anat. Anz. Bd., 23: 384-390.

Essick, Charles R. 1912 The development of the nuclei pontis and the nucleus arcuatus in man. Am. J. Anat., 13: 25-34.

Evans, T. C. 19^+7 Radioautograph in which the tissue is mounted directly on the photographic plate. Proc. Soc. Exp. Biol. Med., 64: 313-315.

Fink, R. P. and Lennart Heimer 1967 Two methods for selective silver impregnation of degenerating axons and their synaptic endings in the central nervous system. Brain, 4: 369. l8l

Fitzgerald, P. J., M. L. Eidnoff, J. E. Knoll, and E. B. Simmel 1951 Tritium in radioautography. Science, 11^: k^b-k^d.

Fliedner, T. M., R. J. Hass, H. Stehle, and A. Adams 1968 Complete labeling of all cell nuclei in newborn rats with H^-thymidine. A tool for the evaluation of rapidly and slowly proliferating cell systems. Lab. Invest., 16: 2^+9-259.

Fox, Clement A. 19^0 Certain basal telencephalic centers in the cat. J. Comp. Neur., 72: 1-62.

Fujita, H. and Setsuyo Fujita 1963 Electron microscopic studies on neuroblast differentiation in the central nervous system of domestic fowl. Z. Zellforsch. Mikroskop. Anat., 60: 463-^-78.

Fujita, Setsuyo 1962 Kinetics of cellular proliferation. Exp. Cell Res., 28: 52-60.

1963 The matrix cell and cytogenesis in the developing central nervous system. J. Comp. Neur., 120: 37-^+2.

196^ Analysis of neuron differentiation in the central nervous system by tritiated thymidine autoradiography. J. Comp. Neur., 122: 311-327.

1965 An autoradiographic study on the origin and fate of the subpial glioblast in the embryonic spinal cord. J. Comp. Neur., 12k: 51-60.

1966 Applications of light and electron microscopic autoradiography to the study of cytogenesis of the forebrain. In: Evolution of the forebrain, phylogenesis and ontogenesis of the forebrain, R. Hassler and H. Stephan (eds.). Thieme, Stuttgart, Germany, pp. I8O-I96.

1967 Quantitative analysis of cell proliferation and differ­ entiation in the cortex of the postnatal mouse cerebellum. J. Cell Biol., 32: 277-287.

Fujita, Setsuyo, M. Shimada, and T. Nakamura 1966 H^-thymidine auto­ radiographic studies on the cell proliferation and differenti­ ation in the external and internal granular layers of the mouse cerebellum. J. Comp. Neur., 128: 191-208. 182

Graupner, Heinz, and Arnold Weissberger 1931 Uber der Verwendung des Dioxanes beim Einbetten mikroskopischer objekte. Mitteilungen zur mikroskopischer Technik I. Zoologiseher Anzeiger, 96: 204-206.

Greenberg, M. L., A. D. Chanana, E. P. Crorikite, L. M. Schiffer, and P. A. Strychmans 1966 Tritiated thymidine as a cybocidal agent in human leukemia. Blood, £8: 851.

Gurdjian, Elisha S. 1925 Olfactory connections in the albino rat, with special reference to the stria medullaris and the anterior commissure. J. Comp. Neur., 38: 127-163.

1927 The diencephalon of the albino rat. Studies on the brain of the rat. J. Comp. Neur., 4-3: 1-114.

1928 The corpus striatum of the rat. Studies on the brain of the rat. J. Comp. Neur., 45: 24-9-281.

Hamilton, J. G., M. H. Soley, and K. B. Eichorn 1940 Deposition of radioactive iodine in human thyroid tissue. Univ. Cal. Publ. Pharmacol., Berkeley, California, pp. 339-368.

Hanaway, Joseph, Jo Ann McConnell and Martin G. Netsky 1971 Histo­ genesis of the substantia nigra, ventral tegmental area of Tsai and interpeduncular nucleus: An autoradiographic study of the mesencephalon in the rat. J. Comp. Neur., 142: 59-74.

Herman, L. and S. L. Kauffman 1966 The fine structure of the embry­ onic mouse neural tube with special reference to cytoplasmic microtubules. Develop. Biol., 13: 14-5-162.

Herrick, C. Judson 1910 The morphology of the forebrain in amphibia and reptilia. J. Comp. Neur., 20: 413-547.

Hicks, Sam P., Constance J. D'Amato, Marie A. Coy, Elaine D. O'Brien, Jean M. Thurston, and David L. Joftes 1961 Migrating cells in the developing nervous system studied by their radiosensitivity and tritiated thymidine uptake. Brookhaven Symp. Biol., l4; 246-261.

Hinds, James W. 1967 Autoradiographic study of histogenesis in the mouse olfactory bulb. Ph.D. Thesis, Harvard University, Cambridge, Massachusetts, 235 pp. 183

Hinds, James W. 1968a Autoradiographic study of histogenesis in the mouse olfactory bulb. I. Time of origin of neurons and neuroglia. J. Comp. Neur., 13^: 287-3014.

1968b Autoradiographic study of histogenesis in the mouse' olfactory bulb. II. Cell proliferation and migration. J. Comp. Neur., 13^: 305-322.

Hinds, James W. and Jay B. Angevine, Jr. 1965 Autoradiographic study of histogenesis in the area pyriformis and claustrum in the mouse. Anat. Rec., 151: ^-56-457.

Hinds, James W. and Trudy Lakin Ruffett 1971 Cell proliferation in the neural tube: An electron microscopic and Golgi analysis in the mouse cerebral vesicle. Z. Zellforsch., 115: 226-26U.

His, Wilhelm looo Zur Geschichte des Gehirns, sovzie der centralen und peripherischen Nerveribahnen beim menschlichen Embryo. Abh. Kon. Sachs. Ges. Wiss., l4: 3^-1-372 (as cited in Keyser 1972).

1889 Die Neuroblasten und deren Entstehungen im embryonalen Mark. Arch. Anat. Physiol, Anat. Abt., _5: 2^9-300.

1890 Die Entweicklung des menschlichen Rautenhirns vom Ende des ersten bie zum Beginn des dritten Monats. I. Verlangertes mark. Abh. Kon, Sachs, Ges. Wiss., 29: 1-7^-.

Holloway, R. L. 1966 Dendritic branching: Some preliminary results of training and complexity in rat visual cortex. Brain Res., 2: 393-396.

Humphrey, Tryphena 1936 The telencephalon of the bat. I. The non-cortical nuclear masses and certain pertinent fiber connec­ tions. J. Comp. Neur. 65: 603.

1963 The development of the anterior olfactory nucleus of human fetuses. In: The Rhinencephalon and related structures. Progress in Brain Research, W. Bargmann and J. P. Schade (eds.), 3: 170-190, Elsevier Pub. Co., Amsterdam.

1966 The development of the human tuberculum olfactorium during the first three months of embryonic life. J. fur Hirnforschung, 9: ^36-^70.

Ifft, John D. 1972 An autoradiographic study of the time of final division of neurons in rat hypothalamic nuclei. J. Comp. Neur., 1^: 193-204. 18k

Joftes, David L. and S. Warren 1955 Simplified liquid emulsion radio- autography. J. Biol. Phot. Assoc., 23: 1*15-150.

Johnston, J. B. 1909 The morphology of the forebrain vesicle in vertebrate. J. Comp. Neur., 19: *+57-539.

1913 The morphology of the septum, hippocampus and pallial commissures in reptiles and mammals. J. Comp. Neur., 23: 371-^78.

1923 Further contributions to the study of the evolution of the forebrain. J. Comp. Neur., 35: 337-^-82.

Kallen, Bengt and K. Valmin 1963 DNA synthesis in the embryonic chick central nervous system. Z. Zellforsch. Mikroskop. Anat., 60: ^91-^96.

Kanemitsu, Akira 1971 Relation entre la taille des neurones et leur epoque d'apparition dans la moelle e'piniere chez le poulet. Proc. Japan Academy, *+7: *t-32-*+37-

1972 Etude quantitative de la neuroge'nese dans la moelle epiniere chez le poulet par 1'autoradiographic. Proc. Japan Academy, 48: 758-763-

Kappers, Ariens, G. Carl Huber, and Elizabeth C. Crosby 1936 The comparative anatomy of the nervous system of vertebrates, including man. Macmillan, New York, 2 vols. (Reprinted i960, Hafner Publ. Co., New York, 3 vols.)

Kauffman, S. L. 1966 An autoradiographic study of the generation cycle in the ten-day mouse embryo neural tube. Exp. Cell Res., k-2: 67-73. ~~

Kershman, J. 1938 The medulloblast and the medulloblastoma. A study of human embryos. A.M.A. Arch. Neur. Psychiat., *4-0: 937-967-

Keyser, Antonius, Jr. 1972 The development of the diencephalon of the Chinese hamster. An investigation of the validity of the criteria of subdivision of the brain. Drukkeri,]' Schippers, Nijmegen, 178 pp.

Knook, Huibrecht L. 1965 The fiber connections of the forebrain; a critical review of the hodology of the telen- and diencephalon with the adjacent mesencephalon, especially with regard to the basal ganglia, based on acute and chronic experiments in the rat. Royal VanGorcum, Ltd., The Netherlands, U59 PP- 185

Konig, Joachim F. R. and Renate A. Klippel 1963 The rat brain. A stereotaxic atlas of the forebrain and lower parts of the brain stem. Williams and Wilkins, Baltimore, 162 pp.

Kopriwa, Beatrix M. and Charles Phillipe Leblond 1962 Improvements in the coating technique of radioautography. J. Histochem. Cyto- chem., 10: 269-284.

Kotzareff, A. 1922 Photographies des organes par 1'emanation du radium ou radiumgraphie (Curiegraphie). J. Radiol. Electrol, 6: 131-133.

Krieg, Wendell, J. S. 1932 The hypothalamus of the albino rat. J. Comp. Neur., _55: 19-89-

Kuhlenbeck, Hartwig and Webb Haymaker 19^9 The derivatives of the hypothalamus in the human brain; their relation to the extra- pyrimidal and autonomic systems. The Military Surgeon, 105: 26-52.

Kupffer, Karl von 1906 Die Morphogenie des Zentralnervensystems. In: Handbuch der vergleichenden und experimentellen Entwicklungs- lehre der Wirbeltiere Hrsg, von 0. Hertwig, Jena (as cited in Keyser 1972).

Lacassagne, Antoine and Jeanne Lattes I92U Methode auto-historadio- graphique pour la detection dans les organes du polonium injecte". Comp. Rend. Acad. Sci. (Paris), 178: 488-4-90.

Langman, Jan, R. L. Guerrant, and N. G. Freeman 1966 Behavior of neuroepithelial cells during closure of the neural tube. J. Comp. Neur., 127: 399-412.

Lauer, Edward W. 19^5 The nuclear pattern and fiber connections of certain basal telencephalic centers in the macaque. J. Comp. Neur., 82: 215-254.

LaVail, Jennifer H. and Merrill K. Wolf 1973 Postnatal development of the mouse dentate gyrus in organotypic cultures of the hippocampal formation. Am. J. Anat. In press.

Lavelle, Arthur and Faith W. Lavelle 1970 Cytodifferentiation in the neuron. In: Developmental Neurobiology, Williamina A. Himwich (ed.), Charles C Thomas, Springfield, pp. 117-164.

Leblond, Charles Phillipe and Katherine Brehme Warren 1965 The use of radioautography in investigating protein synthesis. Academic Press, New York, 3^8 pp. 186

Le Gros Clark, W. E. 1938 Morphological aspects of the hypothalamus. In: The Hypothalamus, W. E. Le Gros Clark, et al. (eds.), Oliver and Boyd, Edinburgh, pp. 1-68.

Levi-Montalcini, Rita 1963 Growth and differentiation in the nervous system. In: The Nature of Biological Diversity, J. Allen (ed.), McGraw-Hill, New York, pp. 261-295-

196^ Events in the developing nervous system,, In: Progress iij Brain Research, Growth and Maturation of the Brain, D. P. Purpura and J. P. Schade (eds.), Elsevier, Amsterdam, k: 1-29-

Levine, Seymour 1962 The effects of infantile experience on adult behavior. In: Experimental Foundations of Clinical Psychology, A. J. Bachrach (ed.), Basic Books, New York, pp. 139-163.

1966 Sex differences in the brain. Scientific American, 21k: 8U-90.

Liquier-Milward, J. 1956 Electron microscopy and radioautography as coupled techniques in tracer experiments. Nature, 177: 619.

London, E. S. 19CA- Etudes sur la valeur physiologique et pathologique de 1'emanation du radium. Arch. Elec. Med., 12: 363-372.

Loo, Yii-Tao 19^1 The mammalian endbrain. I. The septum. Contrib. Biol, Lab. Sci. Soc. China, Shanghai Zool. Ser., 15: 29-68.

Lyser, Katherine M. I96I+ Early differentiation of motor neuroblasts in the chick embryo as studied by electron microscopy. I. General aspects. Develop. Biol., 10: ^33-^66.

1968 Early differentiation of motor ne-aroblasts in the chick embryo as studied by electron microscopy. II. Microtubules and neurofilaments. Devel. Biol., 17: 117-1^2.

Martin A. and Jan Langman 1965 The development of the spinal cord examined by autoradiography. J. Embryol. Exp. Morph., lif: 25-35.

Miale, Irene L. and Richard L. Sidman 1961 An autoradiographic analy­ sis of histogenesis in the mouse cerebellum. Exp. Neur., J+: 277-296. 187

Millhouse, 0. Eugene 19&9 A Golgi study of the descending medial forebrain bundle. Brain Res., 15: 3^1-363.

Morest, D. Kent 1968a Growth of cerebral dendrites and synapses. Anat. Rec., l60: 516.

1968b The growth of synaptic endings in the mammalian brain: A study of the calyces of the trapezoid body. Z. Anat. Entwickl. -Gesch., 127: 201-220.

1969a The differentiation of cerebral dendrites: A study of the postmigratory neuroblast in the medial nucleus of the trapezoid body. Z. Anat. Entwickl. -Gesch., 128: 271-289.

1969b The growth of dendrites in the mammalian brain. Z. Anat. Entwickl. -Gesch., 128: 290-317.

1970 A study of neurogenesis in the forebrain of opossum pouch young. Z. Anat. Entwickl. -Gesch., 130: 265-305.

Moses, Montrose J. I96U Application of autoradiography to electron microscopy. J. Histochem. Cytochem. , 12: 115-130.

NAP 1955 International anatomical nomenclature committee. Nomina anatomica. Revised by the I.A.N.C. appointed by the Fifth International Congress of Anatomists held at Oxford in 1950. Submitted to the Sixth International Congress of Anatomists, held in Paris, July 1955, Baltimore, Williams & Wilkins Co., 1956.

Nauta, Walle J. H. 1956 An experimental study of the fornix system in the rat. J. Comp. Neur., IOU: 2^7-276.

Nauta, Walle J. H. and Webb Haymaker 1969 Hypothalamic nuclei and fiber connections. In: The Hypothalamus, Webb Haymaker, Evelyn Anderson and Walle J. H. Nauta (eds.), Charles C Thomas, Springfield, pp. 136-209.

Niepce de St. Victor 1867 Compt. Rend., 6j>: 505 (as cited in Rogers 1967). ~~

Niimi, Kahee, Ichiro Harada, Yasuhiro Kusaka, and Seiichiro Kishi 1962 The ontogenetic development of the diencephalon of the mouse. Tokushima J. Exp. Med., 8: 203-238. 188

Olds, James F. and Brenda P. Milner 195^- Positive reinforcement produced by electrical stimulation of septal areas and other regions of the rat brain. J. Comp. Physiol. Psych., ^7: kl^-k27.

Olmos, Jose S. De and W. Rex Ingram 1972 The projection field of the stria terminalis in the rat brain. An experimental study. J. Comp. Neur., lU6: 303-333.

Papez, James W. 1929 Comparative Neurology. Thomas Y. Crowell, New York (reprinted in 1961, Hafner, New York, 507 pp.).

1937 A proposed mechanism of emotion. Arch. Neur. and Psychiat., 38: 725-7^3.

Pelc, S. R. 19^7 Autoradiographic technique. Nature, l60: 7^-9-750.

Price, Joseph L. and T. P. S. Powell 1970a An experimental study of the origin and the course of the centrifugal fibers to the olfactory bulb in the rat. J. Anat., 107: 215-237.

1970b The afferent connexions of the nucleus of the hori­ zontal limb of the diagonal band. J. Anat., 107: 239-256.

Purpura, Dominick P. 19&7 Comparative physiology of dendrites. In: The Neurosciences. A Study Program, G. C. Quarton, T. Melnechuk, and F. 0. Schmitt (eds.), Rockefeller University Press, New York, pp. 372-393.

Quimby, Edith H. and Sergei Feitelberg 19&3 Radioactive isotopes in medicine and biology. Basic physics and instrumentation. Lea and Febiger, Philadelphia, 3^-3 pp.

Raisman, Geoffrey, W. M. Cowan, and T. P. S. Powell 1966 An experiment­ al analysis of the efferent projection of the hippocampus. Brain, 89: 83-108,

Rakic, Pasko 1971 Neuro-glia relationship during granule cell migra­ tion in developing cerebellar cortex. A Golgi and electron- microscopic study in macacus Rhesus. J. Comp. Neur., lUl: 283-312.

1973 Kinetics of proliferation and latency between final cell division and onset of differentiation of cerebellar stellate and basket neurons. J. Comp. Neur., 1^7: 523-5^6. 189 Rakic, Pasko and Richard L. Sidman 1968 Subcommissural organ and adjacent ependyma: Autoradiographic study of their origin in the mouse brain. Am, J. Anat., 122: 317-336.

1969 Telencephalic origin of pulvinar neurons in the fetal human brain. Z. Anat. Entwickl.-Gesch., 129: 53-82.

1970 Histogenesis of cortical layers in human cerebellum, particularly the lamina dissecans. J. Comp. Neur., 139: *+73- 500.

Ramon y Cajal, Santiago 1893 La retine des vertebres. La Cellule, 9: 121.

1909-1911 Histologie du systeme nerveux de 1'homme et des vertebres, L. Azoulay (trans.). Maloine, Paris. (reissued 1955? Consejo Superior de Investigaciones Cientificas, Madrid, 2 vols., 1979 PP-)•

1913 Contribution al conocimiento de la neuroglia del cerebro humano. Trab. Lab. Invest. Biol., 11: 255-315.

i960 Studies on vertebrate neurogenesis. L. Guth (trans.), Charles C Thomas, Springfield, U28 pp.

Reichard, P. and B. Estborn 1951 Utilization of desoxyribosides in the synthesis of polynucleotides. J. Biol. Chem., 188: 839-8U6.

Rio-Hortega, P. del 1918 Noticia de un neuvo y facil methodo para la coloracio de la neuroglia y del tejido conjuctivo. Trab. Lab. Invest. Biol., (Madrid) 15: 367-378.

1920 La microglia y su transformation en celulas en bastoni- cito y cuerpos, granulo-adiposus. Trab. Lab. Invest. Biol., 18: 37-83.

Rogers, Andrew W. 1967 Techniques of autoradiography. Am, Elsevier Publ. Co., New York, 335 pp.

Rudebeck, B. 19*+5 Contribution to forebrain morphology in Dipnoi. Acta Zool., 26: 9-156.

Salpeter, Miriam M. 1966 General area of autoradiography at the electron microscope level. In: Methods in Cell Physiology, D. M. Prescott (ed.), Academic Press, New York, pp. 229-253. 190

Sanides, Friedrich 1957 Die Insulae terminales des Erwachsenengehirns des Menschen. J. Hirnforsch., 3: 21+3-273.

Saner, F. C. 1935 Mitosis in the neural tube. J. Corap. Neur., 62: 377-^05. ~

1936 The interkinetic migration of embryonic epithelial nuclei. J. Morph., 60: 1-11.

Sauer, Mary E. and A. C. Chittenden 1959 Deoxyribonucleic acid content of cell nuclei in the neural tube of the chick embryo: Evidence for interkinetic migrations of nuclei. Exp. Cell Res., 16: 1-6.

Sauer, Mary E. and Bruce E. Walker 1959 Radioautographic study of interkinetic nuclear migration in the neural tube. Proc. Soc. Exp. Biol. Med., 101: 557-560.

Sawyer, Charles H. 1967 Some endocrine aspects of forebrain inhibition. Brain Res. 6: U8-59-

Schaper, A. 1897 Die friihesten Differenzierungsvorgange im Central- nervensystem. Arch. Entwicklungsmec'n., 5_: 81-132.

Sechrist, John W. 1969 Neurocytogenesis, I. Neurofibrils, neuro­ filaments and the terminal mitotic cycle. Am. J. Anat., 12h: llU-13^.

Sidman, Richard L. 1961 Histogenesis of mouse retina studied with thymidine-l-l3. in: Structure of the Eye, G. K. Smelser (ed.), Academic Press, Hew York, pp. U87-506.

1970a Autoradiographic methods and principles for study of the nervous system with thymidine-H3. in: Comtemporary Research Methods in Neuroanatomy, Walle J. H. Nauta and Sven 0. E. Ebbesson (eds.), Springer-Verlag, New York, pp. 252-27^.

1970b Cell proliferation, migration and interaction in the developing mammalian ce. tral nervous system. In: The Neuro- sciences; Second Study Program, Rockefeller Univ. Press, New York, pp. 100-107. 191

Sidman, Richard L. and Jay B. Angevine, Jr. 1962 Autoradiographic analysis of time of origin of nuclear versus cortical compo­ nents of mouse telencephalon. Anat. Rec., 1^2: 326-327.

Sidman, Richard L., Jay B. Angevine, Jr., and Elizabeth Taber Pierce 1971 Atlas of the mouse brain and spinal cord. Harvard Univ. Press, Cambridge, 261 pp.

Sidman, Richard L. and Irene Miale 1959 Histogenesis of the mouse cerebellum studied by autoradiography with tritiated thymidine. Anat. Rec., 133: U29-U30.

i960 Pattern of histogenesis in the mouse cerebellum studied by autoradiography. J. Neuropath. Exp. Neur., 19: lk-7.

Sidman, Richard L., Irene Miale, and Wed Feder 1959 Cell proliferation and migration in the primitive ependymal zone; An autoradio­ graphic study of histogenesis in the nervous system. Exp. Neur., l: 322-333.

Sidman, Richard L., Patricia A. Mottla, and Ned Feder 1961 Improved polyester wax embedding for histology. Stain Tech., 36: 2J9~28b.

Skougaard, M. R. and P. A. Stewart 1967 Comparative effectiveness of intraperitoneal and intramuscular 3h_tdR injection routes in mice. Exp. Cell Res., ^5: 158-166.

Smart, I. 1961 The subependymal layer of the mouse brain and its cell production as shown by radioautography after thymidine-H^ injection. J. Comp. Neur., Il6: 325-3^7.

Sprenkel, H. Berkelbach van der 1926 Stria terminalis and amygdala in the brain of the opossum (Didelphis virginiana). J. Comp. Neur., b2: 211-25^.

Stensaas, Larry J. 1967a The development of hippocampal and dorso­ lateral pallial regions of the cerebral hemisphere in fetal rabbits. I. Fifteen millimeter stage, spongioblast morphology. J. Comp. Neur., 129: 59-70.

1967b The development of hippocampal and dorsolateral pallial regions of the cerebral hemisphere in fetal rabbits. II. Twenty millimeter stage, neuroblast morphology. J. Comp. Neur., 129: 71-8^+. 192

Stensaas, Larry J. and Suzanne S. Stensaas 1968 An electron micro­ scopic study of cells in the matrix and intermediate laminae of the cerebral hemisphere of the ^5 mm rabbit embryo. Z. Zellforsch., 91: 3^+1-365.

Taber, Elizabeth 1963 Histogenesis of brain stem neurons studied autoradiographically with thymidine-H^ in the mouse. Anat. Rec., 3J+5: 291.

Taber Pierce, Elizabeth 1966 Histogenesis of the nuclei griseum pontis, corporis pontobulbaris and reticularis tegmenti pontis (Bechterew) in the mouse. An autoradiographic study. J. Cornp. Neur., 126: 219-239.

1967 Histogenesis of the dorsal and ventral cochlear nuclei in the mouse. An autoradiographic study. J. Comp. Neur., 131: 27-51+.

1970 Histogenesis of the sensory nuclei of the trigeminal nerve in the mouse. An autoradiographic study. Anat. Rec., 166:

Taber Pierce, Elizabeth and Ellen Sweet 1967 Histogenesis of deep cerebellar nuclei studied autoradiographically with thymidine-H in the mouse. Anat. Rec., 157• 301.

Trinkaus, J. P. and Marcie C. Gross 1961 The use of tritiated thymidine for marking migratory cells. Exp. Cell Res., 2k: 52-57. ~

Truex, Raymond C. and Malcolm B. Carpenter 1969 Human Neuroanatomy. Williams and Wilkins, Baltimore, 673 pp.

Uzman, L. Lahut 1960a The development of neonatal mouse cerebellum studied by tritiated-thymidine uptake. J. Neuropath. Exp. Neur., 19: 1U6-IU7.

1960b The histogenesis of the mouse cerebellum as studied by its tritiated thymidine uptake. J. Comp. Neur., llU: 137-159-

Valverde, Facundo 1965 Studies on the piriform lobe. Harvard Univ. Press, Cambridge, 128 pp. 193

Valverde, Facundo 1967 Apical dendritic spines of the visual cortex and light deprivation in the mouse. Exp. Brain Res., 3: 337- 352.

Watterson, Ray L., Philip Veneziano, and Alex Bartha 1956 Absence of a true germinal zone in neural tubes of young chick embryos as demonstrated by the colchicine technique. Anat. Rec., 12k: 379*

Weiss, Paul 193*+ In vitro experiments on the factors determining the course of the outgrowing nerve fiber. J. Exp. Zool., 68: 393-^8. ' ~~

1955 Nervous system. In: Analysis of Development, B. H. Willier, P. A. Weiss, and V. Hamburger (eds.). Saunders, Philadelphia, pp. 3^-6-'+01.

White, Lowell E., Jr. 1965 A morphologic concept of the limbic lobe. Intern. Rev. of Neurobiol., 8: 1-3*+.

Willis, Thomas A. l66k Cerebri Anatome. Martzer and Alleftry, London (as cited in White 1965)•

Windle, William F. and M. F. Austin 1936 Neurofibrillar development in the central nervous system of chick embryos up to 5 days incubation. J. Comp. Neur., 63: ^l-^S.

Windle, William F. and R. E. Baxter 1936 The first neurofibrillar development in albino rat embryos. J. Comp. Neur., 63: 173-187.

Yakovlev, Paul I. 1972 A proposed definition of the limbic system. In: Limbic System Mechanisms and Autonomic Function, Charles H. Hockman (ed.). Charles C Thomas, Springfield, pp. 2^1-283.

Young, M. Wharton 1936 The nuclear pattern and fiber connections of the non-cortical centers of the telencephalon of the rabbit (Lepus cuniculus). J. Comp. Neur., 6^: 295~'+01.

Zeman, Wolfgang and James Robert Maitland Innes 1963 Craigie's neuroanatomy of the rat. Academic Press, New York, 230 pp.

Zimmerman, Horst and Julien Fautree (eds.) 1968 Autoradiographic Gesellschaft fur Histochemie Symposion (1967). Fischer, Jena, 383 pp.

Zuckerkandl, Emil 1888 Das Riechbundel des Ammonshorns. Anat. Anz., 3: b25-b3b.