THE CGMPAMTWE ANATOMY OF 'EHE SEREBELLUM QF THE £53553 EUSH BABY (GALAGO flNEGALENSES) AND THE TREE WWW (TUPAM GUS) These: {:cw {he Dances oé m. D. HICRIGAN STATE UNIVERSITY Duane E. Haines 1 96? 1 {1:51. umvsns TYLIHRARI as LIBR ‘4 R Y ”Winn.mmWM 3 1293 1070 Michiganc .JL £21 10 University .Ji}. This is to certify that the thesis entitled TllE COMPARATIVE ANATOMY OF THE CEREBELLUM OF THE LESSER BUSH BABY (Calago Senegalensig) AND THE TREE SHREW (Inpaia Glis) presented by Duane E. Haines has been accepted towards fulfillment of the requirements for Ph.D. degree in Anatomy am A]. Major ole 0r\ Thoma . J nkins Date April 28, 1969 0-169 ABSTRACT THE COMPARATIVE ANATOMY OF THE CEREBELLUM OF THE LESSER BUSH BABY (GALAGO SENEGALENSIS) AND THE TREE SHREW (TUPAIA GLIS) by Duane E. Haines The prosimians are in a key phylogenetic gap between the insectivores and the higher primates, and represent an important 4! group of animals to the anthropologist and neuroanatomist. To the anthropologist the prosimians represent the group of animals from which primate characteristics, both morphologic and behavioral, first emanate. To the neuroanatomist the prosimians, as a group, represent the point from which certain neurological regions become more complex, while other regions undergo regression in the course of primate phylogeny. The cerebellum of prosimian primates has received little attention. It has been only briefly discussed in the course of studies on other portions of the prosimian central nervous system (Elliot Smith, 1903a; Le Gros Clark, 192k, 1926, 1931; Woollard, 1925; Tilney, 1927; Kanagasuntheram and Mahran, 1960; Krishnamurti, 1966). Van Valen (1965) and Campbell (1966a, 1966b) have reviewed the differing opinions on the classification of Yupaia and concluded that a close Tupaia-primate relationship is possibly difficult to establish. The purpose of this study was to describe the gross and micro- scopic anatomy of the cerebelli of the Ghlago and fupaia. An addi- Duane B. Haines tional purpose was to provide information relevant to the "total morphological picture" of Tupaia, and to discuss the phylogenetic implications of the morphology of the Tupaia cerebellum. The anterior lobe of the cerebellum of Tupaia is composed only of a vermal portion, while in Galago large lateral portions of the culmen and central lobule are present. It is possible that the larger paraflocculus of fupaia helps to compensate for the limited development of the anterior lobe. The ansiform lobule of Tupaia is made up of 4 — 5 folia com- pared to 2 folia in the Ghlago. The remaining portions of the hem- isphere are mainly composed of the paramedian lobule which has n folia in both Galago and Tupaia. The cepula pyramidis is vertically fissured in Tupaia and horizontally fissured in Gblago. This has been interpreted as a progressive phylogenetic development. The cytoarchitecture of the cerebellar cortex of the Ghlqgo and Tupaia does not differ significantly from the Mbcaaa (Fox et aZ., 1967). The treeshrew (Tupaia) and the lesser bushbaby (Galqgo) have three cerebellar nuclei on each side, a medial or fastigii, an interpositus, and a lateral or dentate. The medial cerebellar nucleus is separated from the interpositus nucleus by coarse bundles of fibers intermixed with large multipolar neurons. The interpositus and lateral nuclei are incompletely separated from each other through- out their rostro-caudal length. The lateral nucleus of the Galqgo has irregularities in its boundaries that give it the appearance of Duane B. Haines primitive laminations. These laminations are less apparent in Tupaia. All cerebellar nuclei in both animals are composed on small (1n-2u micra) to medium (25—30 micra) sized multipolar neurons. '1‘ THE COMPARATIVE ANATOMY or THE CEREBELLUM OF THE LESSER BUSH BABY (GALAGU SENEGALENSIS) AND THE TREE SHREW (TUPAIA GLIS) by. .- Duane E. Haines A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Anatomy 1969 ACKNOWLEDGMENTS The author wishes to eXpress his sincere thanks to his major Professor, Dr. T. W. Jenkins, fer his guidance and valuable sug- gestions, and to Drs. M. L. Calhoun, J. B. Thomas, and R. E. Carrow for serving on the guidance committee that critically reviewed the manuscript. Appreciation is also due Dr. Daria R. Swindler, honor- ary committee member, for his encouragement to do research on primates. Deep gratitude is expressed to the following investigators for their generous contributions of material to this study: Miss E. Steurwald for one Tupaia gli8_brain, and Dr. Wes Sorenson, University of Missouri, for two Zupaia chinensis. Thanks is also extended to Dr. B. E. Walker for unhampered use of departmental facilities, and generous financial support. The author wishes to express his deep and sincere thanks to Mr. Kenneth R. Holmes, a colleague in GaZago research, for his numerous contributions, both direct and indirect, to this study. Without his cooperation in the GaZago colony since its inception, this study would have been extremely difficult to complete. Appreciation is due Dr. W. W. Armstead, Dean of Veterinary Medicine, and Dr. W. Hoag, Director of CLAR, for funds to establish and maintain the Galago colony. The author extends thanks to Dr. Edward Lauer, University of Michigan, for help with Special bibliographic problems, and to Dr. I L. B. Gerchman for criticisms on portions of the manuscript. ii Special graditude is due Mrs. Dallas Luchsinger, artist, for drawing the gross specimens, and to Mrs. Lester L. Haines, my mother, for her patient and laborous typing of this manuscript. iii VITA ———— DUANE E. HAINES Candidate for the degree of Doctor of Philosophy Final Examination: March 31, 1969 10:00 A.M. Dissertation: The Comparative Anatomy of the Cerebellum of the Lesser Bush baby (Galago senegalensis) and the Tree Shrew (Tupaia glis). Major Subject: Anatomy. Biographical items: Born: May H, 19u3. Springfield, Ohio. Undergraduate studies: B.A. Biology; Greenville College, Greenville, Illinois, 1965. Graduate studies: M.S. Anatomy; Michigan State University, East Lansing, Michigan, 1967. Professional experience: Graduate Assistant in Neuroanatomy, Department of Anatomy, Michigan State University, 1966-1968. Instructor, Department of Anatomy, Michigan State University, 1968—1969. Societies: Member of Beta Beta Beta. Member of the Society of the Sigma Xi. Member of the Michigan Academy of Science, Arts and Letters. iv TABLE OF CONTENTS Introduction . . . . . . . . . . . . . . . . . . Classification . . . . . . . . . . . . . . . . . Galago senegalensis . . . . . . . . . . . . Tupaia gZis . . . . . . . . . . . . . Review of the literature . - . - - - . . . . . . General Remarks - - - - - - . - . - - - - - Cerebellar Cortex-Topography Cerebellar Cortex-Cytoarchitecture - - Cerebellar Nuclei - - - - - . - - . . . . . Materials and Methods Gross Examination - - - - . . - . . - Fixation and Staining - - - - RESULTS AND DISCUSSION Topography of the Cerebellar Cortex - - . - . General Introductory Remarks GaZago senegalensie . - - - - - - . . . Mid-sagittal section - - . - . . . . Anterior lobe - . ~ - - - . . . . Posterior lobe - - - - - . . Figures . . . . . . . Tupaia gZis and Tupaia chinensis - - - - . Mid—sagittal section - ~ - Anterior lobe - - - - - - . - . . . Posterior lobe - - - - Figures . . . o . . . . . . Correlative Discussion ~ - - . . . . - . Flocculus and Paraflocculus - - - - - ~ General Introductory Remarks : - - - ~ . . GaZago senegaZenses o o o . . o o . . . . . Flocculus ' - - . . Paraflocculus ° Copula pyramidis ° Figures ' ° - - . Tupaia gZis and Tupaia chinenSts Flocculus ' ° ° ° - - Paraflocculus ' Copula pyramidis Figures ' ' ° ° ' Correlative Discussion 22 22 24 25 26 28 35 39 39 NO 42 #8 53 61 61 62 62 63 65 69 75 75 76 77 81 86 Cytoarchitecture of the Cerebellar Cortex - General Introductory Granular layer - - - Purkinje layer . - - Molecular layer - - Discussion . - . - - Figures - - - - Cerebellar Nuclei - - - . General Introductory GbZago senegalensis Figures - - . . Tupaia gZis - - - - Figures - - . Remarks Correlative Discussion - Summary and Conclusions . Literature Cited - - - Remarks vi 9H 99 95 99 101 105 107 116 116 120 126 131 135 140 193 146 LIST OF TABLES Table Page 1. Literature reference to the Prosimian para- flocculus and flocculus. . . . . . . . . . . . . . 8H 2. Literature reference to the paraflocculus and flocculus of Insectivores. . . . . . . . . . . . . 84 vii Figure 10. 11. 12. 13. 19. 15. 16. 17. LIST OF FIGURES Schematic classification of the prosimisans used in the present study . . . . . . . . . . . Diagrammatic sagittal view of the prosimian cerebellum as reported in the literature . . . Mid-sagittal view of the cerebellum of the GaZago . Lateral view of the cerebellum of the Galago A semi-diagrammatic drawing of the cerebellar cortex in the Gazago O O O O O O O O O O O I I O O Mid-sagittal view of the cerebellum of the Tupaia . Lateral view of the cerebellum of the Tupaia . . . A semi-diagrammatic drawing of the cerebellar cortex of the Tupaia . . . . . . . . . . . . . . . Anterior view of half of the cerebellum of the Tupaia I O O O O O O O O O O O O O O O O O O O 0 Lateral view of the flocculus and paraflocculus of the Gazago O O O I O O O I O O O I O I I Anterior view of the flocculus and paraflocculus of the Galago . . . . . . . . . . . . Dorsal and ventral views of the paraflocculus of the GaZago . . . . . . . . . . Caudo-ventral view of the left cerebellar hemisphere of the Galago . . . . . . . . . . . . Dissected caudo—ventral view of the left cerebellar hemisphere of the GaZago . Caudo—ventral view of the right cerebellar hemisphere of the Galago . . . . . . . . . Lateral view of the flocculus and paraflocculus of the Tupaia . . . . . . . . . . . . . . . . . . . Dorsal and slightly lateral view of the paraflocculus of the Zupaia . . . . . . . . . . . . viii Page 11 36 37 38 49 50 51 52 7O 7O 71 72 73 7M 82 83 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. Caudo-ventral view of the right half of the cere- bellum Of the mpaia 0 O O O O O O O O O O O O Dissected caudo-ventral view of the left half of the cerebellar hemisphere of the Tupaia . . . Cytologic layers of the cerebellar cortex of the Gazago and Q‘upaia O O O O O I O O O I O O O O Photomicrographs of the Purkinje cell, Golgi cell, and cerebellar islands. . . . . . . . . . . . Photomicrographs of granular cells with long and short dendrites . . . . . . . . . . . . . . Photomicrographs of a low magnification and high magnification of a mossy fiber rosette . . . . . Photomicrographs of the granular layer of the Gazago and mpaia O O O O O O O O O O O O O O O O Photomicrographs of granular cell axons . . . . . Tracing of granular cell processes . . . . . . . Diagrammatic illustration of the relative position of the Golgi neuron and Purkinje cell . Tracings and photomicrographs of Purkinje cell processes . . . . . . . . . . . . . . . . . Diagrammatic illustration of the relationship of parallel fibers and climbing fibers . Photomicrograph and tracing of the intermediate cell of Lugaro . Lower power photomicrograph of the molecular layer. Photomicrograph of the upper and lower portions of the molecular layer. . . . . . . . . . . . . . Tracings and photomicrographs of superior, middle and inferior stellate cells Photomicrographs of typical stellate cells ix 8% 85 107 107 108 108 109 109 110 110 111 112 112 113 113 119 115 35. 36. 37. 38. 39. HO. #1. #2. 43. an. ”5. 46. Q7. 98. H9. 50. Photomicrographs of the Purkinje cell pericellular. Tracings of approximately the rostral one-third of the cerebellar nuclei in the Galqgo . . . . . . Tracings of approximately the middle third of the cerebellar nuclei in the Ghlago . . . . . . . . Tracings of about the caudal third of the cerebellar nuclei in the Gblqgo . . . . . . . . . . A diagrammatic dorsal view of the cerebellar nuclei drawn to scale . . . . . . . . . . . Photomicrograph and tracings of multipolar cell bodies of the medial cerebellar nucleus . . . Photomicrograph and tracings of multipolar cell bodies of the interpositus nucleus . . . . . . . Photomicrograph and tracings of multipolar cell bodies of the lateral cerebellar nucleus . . . . Tracings of about the rostral third of the cerebellar nuclei in the Tupaia . . . . . . . . . Tracings of about the middle third of the cerebellar nuclei in the Tupaia . . . . . . . . . . Tracings of about the caudal third of the cerebellar nuclei in the Tupaia . . . . . . . . . A diagrammatic dorsal View of the cerebellar nuclei of the Tupaia . . . . . . . . Photomicrographs of the multipolar cell bodies of the medial cerebellar nucleus Photomicrographs of the multipolar cell bodies of the neurons of the interpositus nucleus Photomicrographs of the multipolar cell bodies of the neurons of the lateral nucleus . . . . . . . A diagrammatic dorsal view of the cerebellar nuclei of Galago and Tupaia drawn on the same scale. . . . . . . . . . . . . . . . . . 115 126 127 128 129 129 130 130 135 136 137 138 138 139 139 1u2 INTRODUCTION An understanding of the function of a part must be preceded by a complete understanding of the morphology of the part. Com- parative neuroanatomical studies provide important information on structures and systems of structures that have either become more complex or undergone regression in the evolution of higher primates and man. A broad understanding of a wide range of animals is necessary when attempting to trace phylogenetically the development or regression of a structure or system. In the nervous system comparative studies provide a great deal of information relative to similar systems in higher and lower forms. The prosimian primates are at a key phylogenetic position not only from the an— thropologist's point of View, but also from the neuroanatomist's standpoint. To the anthrOpologist the prosimian represents the link from insectivores to higher primates and man. To the neuroanatomist the prosimians represent that point from which certain regions begin to evolve (e.g., cerebral cortex) and certain other regions become less important to the function of the organism (e.g., olfactory system). Since the preliminary studies of Elliot—Smith (1903a), Le Gros Clark (1924, 1926, 1931), Woollard (1925) and Tilney (1927) very little attention has been focused on the cerebellum of prosimian primates. It is the goal of this study to provide a substantial foundation for future studies on the prosimian cerebellum by precise- ly defining the gross and microscopic anatomy of the cerebelli of the two prosimian primates, Galago and Tupaia. 1 A second and by no means minor fact of consideration is the author's interest in physical anthropology and evolution. One of the animals used in this study (Tapaia) has received a great deal of attention concerning its phylogenetic classification. The question is, simply stated, whether it should be classified as an Insectivore or a Prosimian. This study does not propose to answer this question in its entirety, but to provide more detailed information to aid in a clearer understanding of the "total morphological picture" as its importance is stressed by Le Gros Clark (1959). This study centers on the comparative anatomy of the cerebelli of Tupaia glis, a low prosimian, and GaZago senegalensis, a more advanced prosimian. Com- parative studies on the nervous system contribute a great deal to current anthrOpological problems, as pointed out by Holloway (1968). "Two aspects of neurology have particular significance for the anthropologist's area of study: palaeoneurology and comparative neurology. The former studies the changes as they appear in the fossil record, while the latter attempts to better understand the nature of extant forms through study 'of structure and function." Animals of the type used in this study, though not rare, are difficult to obtain. A colony of GaZago senegalensis was established in the Anatomy Department in 1967 (Holmes et aZ., 1968), and animals from this colony were used in the present study. The cerebellum was chosen because of its relationship to agility and behavior and the phylogenetic implications of this relationship in prosimians (Stephan and Andy 1964), and because of the voids in the literature on the details of the prosimian cerebellum. The early investigators previously mentioned, and recent investigators, deal with the prosimian cerebellum only in passing. With the current trend to use small primates in the laboratory, precise data on basic anatomy are surely needed. Without this basic anatomical information more advanced studies cannot be initiated, or if they are, the relative lack of information imposes an ever present obstacle. The author does not consider this study an end in itself, but hopes it provides substantial foundation for future anatomical, physiological and anthropological studies on the prosimian cere- bellum, and indeed the entire prosimian brain stem. The potential that a small primate possesses for the medical researcher, and the importance of the prosimians to the anthropologist, warrant clear descriptions and definitions of basic anatomical parameters. CLASSIFICATION All living primates are grouped into two large suborders, the Prosimii and the Anthropoidéa. This present study deals with members of the suborder Prosimii, and incorporates a borderline prosimian, the tree shrew, and a more advanced prosimian, the lesser bushbaby. The suborder Prosimii is further subdivided into six families that span the gap in the evolution of primate morphology between the insectivores and Simians (Buettner-Janusch 1966; Napier and Napier 1967). These families are the Tupaiidae, Lemuridoe, Indriidae, Daubentoniidae, Lorsidbe and Threiidae, in respective order from insectivores to primates (Fig. l). GMLAGO SENEGALENSIS The lesser bushbaby (Galago senegalensis) is a member of the family Lorisidae, therefore represents a more advanced prosimian. The complete classification of Galago is as follows (Fig. 1): Order Primates Suborder Prosimii Infraorder Lorisiformes Subfamily Galaginae Genus Galago Species Senegalensis TUPAIA GLIS The tree shrew (Tupaia gZis) is a member of the family Thpaiidae. The relative phylogenetic position of Tupaia has been investigated by numerous anatomists and anthropologists and even to date opinion differs. Van Valen (1965) reviewed early and recent literature and comes to this conclusion: mwmcocfico mammse 9.3m wwmaseV meow amaze omowflmms omoMRDqu oopwwsoaH mmEQOMMQSqu ompwwoouoondma «wEHmoem imovmerum mwmcoammeoom owmamo meowwmamw oopfimwnoq moenomflmwnoq omowfimpme moepomwmnme newcomm modem . mawemmndm hawamm noonomnwcH geopoosm noose .hoSpm pcomona opp CH poms mcmHEwmoom may mo coflumOMMHmmmao OHpMEonom H .mwm "I believe the better course...is to refer the Tupaiidae to the Insectivore without taxonomically expressing a special relationship to any of the other groups of insectivores." Campbell (1966a, 1966b) re-evaluated the evidence of the nervous systems and also concludes that a close tupaiid-primate relationship is difficult to establish. It is significant to note, however, that all recent studies on the tupaiid nervous system have been restricted to the pyramidal tract (Jane at aZ., 1965; Campbell, 1965; Verhaart 1966), or to observation on the visual system (Tigges, 1964, 1966; Tigges et al., 1967; Campbell et aZ., 1966; Glickstein et al., 1966; Glickstein, 1967; Snyder and Diamond, 1968). It does not seem valid to definitively suggest a phylogenetic position for Yupaia based on discrepencies in two neurological systems. There are other systems just as important as these, and this position is open for challenge. Stephan and Andy (1969) briefly commented on the "larger cerebellum (in Tupaia) than either of the monkeys, ChZZithrix and Leontocoebus." They further state that the size of the GaZago cerebellum may be related to the "marked agility" of the lesser bushbaby. In this present study the cortical topography of the cerebellum of Tupaia is just as complex as that of Galago, even though the latter cerebellum is slightly larger. If the suggestion of Stephan and Andy (1969) is partially true, the complexity of the Tupaia cerebellum should be ex- plained, since this animal is not necessarily noted for its agility. In view of varying opinions concerning Thpata claSSification, this author has consulted the evaluations of recent authorities (Le Gros Clark, 1959; Buettner-Janusch, 1963, 1966). Since an animal is classified on the basis of all its morphological character- istics and affinities for a particular family, Le Gros Clark (1959) has stated for Tupaia: "...they show in their total morphological pattern such a remarkable number of in- timate resemblances to the lemurs, and particularly the Lemuriformes, that their affinities with the latter can no longer be doubted." The tree shrew (Tupaia gZia) is undoubtedly a primitive primate, but classified as a primate (Buettner-Janusch, 1963; Napier and Napier, 1967). The full schema of Tupaia classification is as follows (Fig. 1): Order Primates Suborder Prosimii Infraorder Lemuriformes Family Tupaioidea Subfamily Tupaiinae Genus Tupaia species glis & chinensis REVIEW OF THE LITERATURE GENERAL REMARKS The cerebellum, as a general subject, has received varied amounts of attention depending on the animal or group of animals under consideration (Ariens Kappers et al., 1936). In the first part of a comprehensive work on the comparative anatomy of the cere- bellum Larsell and Jansen (1967) studied vertebrate animals covering the phylogenetic scale from hagfish and lampreys to birds. A large amount of information has been published on the cerebelli of higher mammals and man. The animals most commonly used in physiologic studies on the cerebellum are cats and various simian primates (Eccles et aZ., 1967). Therefore, a great deal of recent information is available for the cat (Snider, 1990; Larsell, 1953; Flood and Jansen, 1961; and others) and for various simian primates (Braiten- berg and Atwood, 1958; Voogd, 1967; Mussen, 1967; and others). These studies will not be reviewed since these animals possess cere- belli that are considerably more complex than the prosimian cerebellum, cats are not primates, and a review of these studies would not con- tribute to the present study. Future reference to the cerebelli of more advanced forms, e.g., cat and simian, will be utilized only when it helps to clarify a point. CEREBELLAR CORTEX-TOPOGRAPHY A thorough review of the literature has revealed very little available information on the detailed morphology of the prosimian cerebellum. The early studies of Le Gros Clark (192#, 1926, 1931), Elliot Smith (1903a), Woollard (1925) and Tilney (1927) have been followed by the recent observations of Kanagasuntheram and Mahran (1960) and Krishnamurti (1966). In all of these cases the cerebellum was not the primary topic of discussion and consequently received a parcity of attention. In the following review only those articles on the prosimian nervous system will be considered. Other animals with a morphologically simple cerebellum (e.g., rat, small marsupials monotremes, avian, etc.) will be included only when they clarify and/or compliment a particular point of the discussion. Elliot Smith (1899b) in an early study on the brains of sloths and anteaters noted that the terminology used in human anatomy to describe the morphology of the cerebellum is ill-adapted to com- parative observations, particularly when dealing with cerebelli of simple organization. In a later study of the lemur brain, Elliot Smith (1903a) proposed a schema based on the recognition of a "pre— clival" fissure. This fissure is the deepest fissure crossing the mesial line, and is universally present in all mammals (Elliot Smith, 1903a). Using this fissure as a landmark Elliot Smith (1903a) desig— nated three main lobes: The anterior lobe (rostral to the primary fissure - a term synomous with "preclival"), the middle lobe (between the primary fissure and "the shallower fissura secunda"), and the posterior lobe (caudal to the secondary fissure). This concept is diagrammatically illustrated in figure 2. The criterion for the term primary fissure is clear, however, the criterion for the term secondary 10 fissure is not clear. A review of the literature on this matter has led this author to the conclusion that this term (secondary fissure) is a type of regressive phylogenetic designation. This means that the fissure was present in more complex cerebelli and its prescence was regressively traced (phylogenetically) to the very simple forms. Larsell (1952), in his monumental work on the cere— bellum of the white rat, noted that a prominent fissure secunda was present in the 21 day fetus. Aciron in 1950 (Larsell, 1952) further noted that this fissure is distinct in the 35 mm rat embryo. Both of these authors also noted that other fissures were present in these early stages, therefore the possibility of embryological se— quence would not necessarily hold, and consequently the term fissure secunda is open to criticism. Elliot Smith (1903a) also distin- guished a band of cerebellar cortex that joined the paraflocculus to the pyramidal lobe of the vermis. On this problem he stated: "...it is desirable to have some term with which to distinguish the band of cortex linking the pyramid to the paraflocculus. I shall therefore call it the "copula pyramidis"." The cerebelli of the tree shrews Tupaia minor and Ptilocercus Zowi (pentailed tree shrew) have been briefly mentioned by Le Gros Clark (1929, 1926). In Inpaia minor the three main lobes and the two main fissures were noted (Fig. 2), a distinct c0pu1a pryamidis was present, and the anterior surface of the cerebellum was indented by the tectum of the mesencephalon. The copula pyramidis appeared to be "notched transversely" close to the paraflocculus, and no FPcul FPcen Fig. 2. PP FPr A B ., Diagrammatic sagittal view of the general organization of the prosimian cerebellum as reported in the literature. ABBREVIATIONS lingula A— anterior lobe central lobule culmen median lobule B“ middle lobe pyramidal lobule . uvula C— posterior lobe nodulus FP — primary fissure FPcen. - precentral fissure FPcul. - preculminis fissure FPn. - postnodular fissure FPr. - prepyramidal fissure FS — secondary fissure 11 12 distinct connection between the flocculus and nodulus was noted. (Consult table I for the review of flocculus and paraflocculus.) In.PtiZooercuZ lawi (Le Gros Clark, 1926) the cerebellum is simpler in tapography than Tupaia. The three main lobes and two main sulci are present. He further divided the anterior lobe into a lingula- central lobe region and a culmen by the fissure preculminis (Fig. 2). The middle lobe consists of the median lobule and pyramidal lobule, which is joined by a simple cepula pyramidis. Behind the fissure secunda is the uvula and nodulus. In the Tupaia and Ptilocercus (Le Gros Clark 1929, 1926) the anterior and middle lobes have a varied number of short shallow sulci. The lateral hemispheres, which are principally lateral extensions of the median portion of the mid- dle lobe, also have a varied number of sulci. The cerebellum of Threius has been discussed by Elliot Smith (1903a), Woollard (1925) and Tilney (1927). Elliot Smith (1903a) noted that the topography of cerebellar cortex was more complex than either insectivora or marsupialia. The three main lobes are present and as noted by Tilney (1927), sulci rarely crossed from the vermis onto the lateral hemisphere. Woollard (1925) also noted three main lobes separated from each other by the fissure secunda and fissure prima. At this point it should be noted that the term fissure secunda begins to create confusion. Woollard (on marsiue) states: "The fissura secunda or sulcus pre- pyramidalis lies on the sloping posterior surface of the cerebellum just in front of the pyramid." 13 This statement coupled with a careful study of the plates indicates that the secondary fissure of Elliot Smith (1903a) and the secondary fissure of Woollard (1925) are not the same. Woollard (1925), con- sequently, considers the pyramid as a component part of the posterior lobe. The relative position of the sulcus prepyramidalis of Woollard (1925) is indicated on figure 2. The medial and lateral portions of the middle lobe are creased by a variable number of sulci, and both the lingula and nodulus'are hidden from view (Elliot Smith, 1903a; Woollard, 1925). The differing Opinions on the relative size of the flocculus and paraflocculus are noted in table I (p. 84). Kanagasuntheram and Mahran (1960) briefly discussed the cere- bellum of the lesser galago (Galago senegalensis senegaZensis). They note a primary fissure, a secondary fissure, and the respective an- terior middle and posterior lobes. The secondary fissure in this case is caudal to the pyramid of the vermis. The paraflocculus is associated with the pyramid of the vermis by the copula pyramidis which, accord- ing to these authors "...is hidden from the surface by the lateral portions of the middle lobe." A fissure preculminis is also noted (Fig. 2). Their fissure preculminis is caudal to the culmen and is termed fissure precentralis (Fig. 2) by other authors. The obser- vations on the flocculus and paraflocculus are noted on Table I. Krishnamurti (1966) reported on the cerebellum of the slow loris (Nycticebus coucang coucang), and noted several basic differences. The greatly enlarged anterior lobe is divided into a culmen, central lobule and lingula by the fissures preculminis and precentralis iii. 1' 14 respectively. He also designated a prepyramidal fissure on the middle lobe and a secondary fissure separating middle and posterior lobes. Krishnamurti (1966) also noted that the copula pyramidis is almost completely hidden from view by the lateral extensions (hemis— pheres) of the middle lobe. The observations of the flocculus and paraflocculus are noted in table I. It is appropriate at this point to clarify a conflict in the literature. The central lobe of Krishnamurti (1966) in the slow loris is between the culmen, which is directly rostral to the fissure prima, and the lingula. The central lobe of Kanagasuntheram and Mahran (1960) in the lesser galago is designated as that region directly rostral to the fissure prima. In other words, these authors have designated these two areas exactly opposite. This author, in view of this variability, has consulted the opinion of Crosby et al., (1962) who also point out that the terminology applicable to the human cerebellum is not applicable to subhuman and subprimate forms. They further note that to the comparative neurologist the primary fissure is the primary landmark, and the main subdivisions of the anterior lobe are culmen, central lobule, and lingula in respective order from fissura prima to the anterior medullary velum. Since most investigators of the prosimian cerebellum refer to the culmen as the region directly rostral to the fissure prima, figure 2 has been drawn, to conform to this general consensus. Le Gros Clark (1931), in a study of the brain of the mouse lemer (Microcebus murinus), noted that the cerebellum was simple and 15 resembled the same structure in Threius. A primary fissure and secondary fissure are present, and the corresponding 3 main lobes. The lateral lobes are small in relation to the vermis, and aside from the primary fissure only one shallow sulcus crossed from the vermis onto the hemisphere. The dispensation of the flocculus and paraflocculus can be noted in table I (p. 87). The cerebelli of insectivores are very simple in structure and the terminology applied to prosimians is also applicable to this animal group. Elliot Smith (1902b) and Le Gros Clark (1928, 1932) have reported on the_general anatomy of the cerebellum in a variety of insectivores and in the elephant shrew. In all cases a fissure prima and fissure secunda were present with the respective main lobes. The inconclusive observations on the flocculus and paraflocculus in this group are noted in table II. Le Gros Clark (1932) applied the term "post nodular fissure" to the deep groove between the uvula and the nodulus (dotted line - figure 2). This terminology has not been used in relation to any prosimian, only to insectivores. The lateral hemispheres of the prosimian cerebellum are primarily lateral estensions of the middle lobe. Elliot Smith (1902a, 1903a, 1903b) designated three separate portions of the hemisphere as areas A, B, and C from rostral to caudal. The dopula pyramidis was occasion- ally referred to as area D. He described area A as the single folium caudal to the primary fissure; area B as the most lateral portion of the hemisphere and possessing a very narrow connection to the middle lobe; and area C as that portion between area B and the c0pula 16 pyramidis (area D). These regions and divisions are acknowledged for Tupaia minor and Ptilocercua Zowi (Le Gros Clark, 1924; 1926) however no other investigator on the prosimian nervous system dis- cussed these divisions. The available information on the hemispheres of the prosimian cerebellum is extremely vague. Other reports of animals with a topographically simple cere- bellum include those of Obenchain (1925) on small South American marsupials, Larsell (1952) and Zeman and Innes (1963) on the rat, and Dillon (1962) on monotremes. Pertinent studies on the cere- bellum of the mouse include those of Miale and Sidman (1961), and Haddara and Nooreddin (1966). CEREBELLAR CORTEX-CYTOARCHITECTURE The cytoarchitecture of the cerebellar cortex has been investi— gated in a wide variety of mammalian and submammalian forms. Larsell and Jansen (1967) reviewed the literature on the vertebrates ranging from Hagfish to birds. The cytoarchitecture of the mammilian cere— bellum has been studied by normal anatomical methods and by various degeneration techniques. Fox et al., (1967) and Eccles et aZ., (1967) have reviewed the pertinent literature, and the following brief description is from their works. The cerebellar cortex is made up of an outer molecular layer, a row of Purkinje cells, and an inner granular layer. The molecular layer is made up of two main cell types, the superior stellate cell and the inferior stellate cell, or more commonly called the basket cell. The superior (or outer) stellate cells come into synaptic relationship with the dendritic 17 field of the Purkinje cell, and the basket cell (inferior stellate) sends out a series of long processes that surround the cell body of the Purkinje cell. The Purkinje cell is the main efferent neuron of the cortex. Its dendritic field is compartmentalized upward in- to the molecular layer, and its single axon (with or without collat— erals) descends from the underside of the cell body into the white matter. The granular layer contains the small (5-8 micra) granular cells with their characteristic claw-like terminations on the mossy fibers, and the larger (large stellate neurons) Golgi neurons. The granular cells in their terminations on the mossy fibers form the cerebellar or glomerular islands. The afferent fibers of the cortex are the mossy fibers and the climbing fibers. They have origin from the deep cerebellar nuclei as well as various brain stem regions (Eccles et aZ., 1967) and are generally demonstrated by staining for degenerating fibers following specifically localized lesions. A thorough review of the literature has revealed no available information on the cytoarchitecture of the cerebellar cortex for any prosimian primate. Due to certain consistent mammalian characheristics it is reasonable to assume that the prosimian cerebellar cortex will, in its general architecture, conform to that of the rhesus monkey (Fox at al., 1967). CEREBELLAR NUCLEI It is well known that the cerebellum of man contains eight cere- bellar nuclei, (Crosby et al., 1962). In the anthropoid apes, and in the higher simian primates four morphological nuclei can be demon- 18 strated on each side (Ariens Kappere at al., 1936; COOper and Cour- ville, 1968). It is equally well known that the cerebelli of dogs and cats contain three nuclei on each side (Flood 8 Jansen, 1961; Singer, 1962). In the lower Mammals (monotremes, marsupials, etc.) there are only two nuclei on each side, a roof nucleus and a lateral nucleus (Ariens Kappers at aZ., 1936). No information could be found on the gross appearance or cytoarchitecture of the cerebellar nuclei in any prosimian primate. Obenchain (1925) in her study of small South American marsupials commented: "The deep nuclei form a pair of large oval masses almost meeting in the mid— line... there are only slight indications of differentiations into separate nuclei (dentate and roof nuclei)." Tilney (1927) commented briefly on the cerebellar nuclei of Thraius. He stated that they resemble the cerebellar nuclei of other low mammalian forms, and that they lack the size, definition and configuration of higher forms. However he does not have illus- trations for his Opinion, and he does not reference the "other low mammalian forms." MATERIALS AND METHODS GROSS EXAMINATION The cerebelli of nine adult lesser bushbabies (Galago senegalensis), and five adult tree shrews (Tupaia glis) were used in this study. The brains of two adult Tupaia chineneis also became available for this study. The T. chinensis were immersion-fixed (by the contributor) in formalin-Alcohol-Acetic acid without the skull being opened. Con— sequently, these two brains were used only in the gross portion of this study since they were inadequate for histologic study. The ‘gross features of these two cerebelli were quite clear (folia, sulci, etc.) and subsequently contributed a great deal to the comparative value of this study. Some of the brains (in G. senegalensis and T. glis) were fixed by perfusion of the animals, while others were carefully removed and fixed by immersion in the apprOpriate fixative. The brain stem was transected at the level of the caudal mesencephalon and the cerebelli left attached to the pons and medulla to facilitate gross examination. Gross examinations of the cortex were made bilaterally under a dissecting microscope using standard micro—dissecting instruments. Two dissecting scopes were used, a standard model, and one having a zoom lens. The latter allowed fine observation of the more discreet macroscopic points. In addition, drawings of the topography were made to the scale of 1 cm = 1 mm using calipers and a metric rule. All drawings were repeatedly checked for accuracy on horizontal, vertical and oblique angles. 19 20 FIXATION AND STAINING A variety of histological techniques were used to illucidate the cytoarchitecture of the cortex and cerebellar nuclei. Cerebelli that had been perfusion-fixed and immersion-fixed were used for the histologic portion of this study. A variety of fixatives were used, depending on which stain or special technique was to be applied to a specific sample of tissue. Fixation fluids for each stain or stain procedure used in this study. Fix Procedure 80% Alcohol Thionin 10% BNF * Einarson method 10% BNF * Kluver and Barrera 15% BNF * Cajal Method IV 100% Alcohol + 1-3 drop ammonia Cajal Method III chloral hydrate-5-6gm 100% Alcohol -25cc HOH -75cc Cajal Method VI 5% potassium dichromate -2pt 1% Osmium tetroxide -1pt Rapid Golgi * BNF = Buffered Neutral Formalin The luxol fast blue-cresylecht violet method of Kluver and Barrera (1953) the gallocyanin method Einarson (1932), and a standard 21 thionin method (1959)1 were used as Nissl stains. The luxol fast blue- cresylecht violet also served as a standard fiber-Nissl combination. Sections stained with the preceeding methods were cut at 8, 10, 12 and 20 micra on a rotary microtome. Silver stains were employed to illucidate the cytological details of the cortex and cerebellar nuclei. The standard rapid-Golgi method (Conn at al., 1962) was used, as were three methods of the Cajal reduced silver technique, Cajal Method III, IV, and VI (Jones, 1964). The specific fixative for each method is noted in table 3. The silver methods utilized in this study were block stains, and followed by dehydration and embedding in paraffin. The silver blocks were subsequently sectioned at 20, 25, 40, 50, 60, 70, 80 and 100 micra, mounted, coverslipped and stored in a dark place. The drawings of the cerebellar nuclei were made with the aid of a Prado Universal projector2 to insure accuracy. Every 10th or 20th section (indicated in each figure) was used and traced to show the progressive macroscopic appearance of the nuclei from rostral to caudal. The tracing of the nuclei was done on Luxol fast blue- cresylecht violet sections and confirmed on the gallocyanin stain. 1. Clinical Laboratory Procedures: Pathological Anatomy Technique. U. S. Naval Medical School, Dept. of Navy. 1959. 2. Leitz Co., Germany. RESULTS AND DISCUSSION TOPOGRAPHY OF THE CEREBELLAR CORTEX GENERAL INTRODUCTORY REMARKS The criterion for dividing the cerebellum into main regions or lobes is not well established. Elliot Smith (1903a) suggested an anterior, middle and posterior lobe based on the recognition of a preclival (prima) fissure and a secondary fissure (fissura secunda). Bradley (1904, 1904-05) in an apparent effort to clarify the problem introduced the following method: "In order to avoid confusion, no pre- existing names were applied to these lobes and fissures, but the simplest method of designating them-~that of letters and figures-—was employed. The five lobes (...of the simplest workable type of cerebellum...) were called A, B, C, D, and E, commencing the enumeration with the most anterior, or rather with the one nearest to the anterior medullary velum. The fissures were similarly designated as I, II, III, and IV." ‘ Comparing the methods of Elliot Smith (1903a) and Bradley (1904) one sees that the primary fissure and secondary fissure of Elliot Smith are fissure II and sub-fissure d respectively of Bradley. Anterior lobe A is area A 8 B, middle lobe B is area C, and posterior lobe C is area D 8 B, respectively, of each author. In the rat (Larsell, 1952) and avian (Larsell and Jansen, 1967) the method of numbering the folia from I-X has been utilized. 23 23 This designation resulted in the following classification: I - lingula II - ventral lobule centralis; sublobules a, b III - dorsal lobule centralis; sublobules a, b IV - ventral lobule culmenis V - dorsal lobule culmenis VI - declive; sublobules a, b, c, d VII - tuber; sublobules a, b VIII - pyramis IX - uvula; sublobules a, b, c, d X - nodulus Crosby et aZ., (1962) revert to a more basic and possibly more logic designation by suggesting only two main lobes; an anterior lobe rostral to the primary fissure, and a posterior lobe encompassing all portions caudal to the primary fissure. The divisions of Elliot Smith (1903a), in one form or another. have been used by all investigators on the prosimian cerebellum. The methods of Bradley (190%, 190H-05) are rarely used and lead the reader into a great deal of confusion. The method used by Larsell (1952) for the rat has not been applied to the prosimian cerebelli; and since the primary fissure is the main landmark, the division of Crosby et al., (1962) is certainly the most logical. The primary fissure is the only consistently agreed upon land- mark in all prosimian cerebelli, and all other simple mammalian cere- belli. It is deepest of all fissures, the only (or main) one to cross from the vermis into the hemisphere in the adult, and it appears early in the embryological development. Using this fissure as a landmark, it is suggested that all main fissures be named in order rostrally and caudally from this point. Rostrally there is a precentral fissure and a postcentral fissure (prelingual), and 24 caudally a prepyramidal fissure, postpyramidal fissure, and a prenodular fissure. This terminology employs the only agreeable landmark, designates the fissures in order rostrally and caudally in relationship to the main regions of the vermis, and is entirely applicable to both Galago and Tupaia (Fig. 3 and 6). It is further suggested that this designation of main fissures be used in all prosimian cerebelli because of a general lack of a basic criterion for naming fissures in the adult animal. The above mentioned terms will be used throughout the remainder of this study. The method used by Larsell (1952) for the rat for designating the vermal folia I-X is partially applicable to the prosimian cere— bellum and modifications of his method will be used. It is extremely confusing to designate gyri and sulci by only upper or lower case letters or numbers (illustrated by reading Bradley), therefore all folia or groups of folia will be named. The best and most applic- able term presently applied to any Specific region will be retained, and for those areas not presently named, a justifiable term will be drawn from the comparative literature. GALAGO SENEGALENSIS The lesser bushbabies used in this study are adult, therefore all discussion is of the adult form. No attempt is made to suggest embryological origin of any part, since this would be pure conjecture. The lesser bushbaby (GaZago senegalensis) is found throughout most of Africa south of the Sahara (Hill, l953). The most striking behavioural characteristic of the Galago is its mode of locomotion. 25 They move by leaping on their hind limbs, the leap being as much as 10 feet in an oblique direction (Hill, 1953) or 7 feet, 4.75 inches in a vertical direction (Hall-Craggs, 1965). This mode of progression requires this animal to have a keen sense of forward and backward balance. When not in a stress situation the Ghlago will proceed in a quadrapedal manner (personal observations), however the usual mode of locomotion is via a hOpping attitude. They also demonstrate a good sense of balance by assuming an entirely erect posture, from a squatting position, with great rapidity. Mid-Sagittal Section A mid-sagittal view of the Galago cerebellum (Fig. 3) illustrates a structure of moderate complexity. The primary fissure is the deepest fissure, and has within its bounds three main folia on its rostral and caudal aspects. The sulci within the primary fissure are continuous from the vermis into that portion of the fissure that extends into the lateral hemisphere. This is the case only for the primary fissure, the sulci in all other main fissures are restricted to the vermal portion of their respective lobes and have no contin- uation into the hemisphere. The primary fissure divides the cere— bellum, on mid-sagittal view, into two main lobes of almost equal size. The anterior lobe rostral to, and the posterior lObe caudal to the primary fissure. The nodulus of the posterior ldbe is technically part of the flocculonodular complex. 26 Anterior lobe The anterior lobe is divided by a precentral fissure and a poet- central fissure (prelingual fissure) into a culmen lobule, central lobule and lingual lobule respectively (Fig. 3, 5). When the cere— bellum is viewed from its lateral aspect (Fig. u) it is noted that the primary fissure is continuous into the hemisphere, consequently the culmen.has a hemispheric portion. It is further noted that the precentral fissure is also continuous into the hemisphere subse- quently giving rise to a hemispheric portion of the central lobule (Fig. 5). The hemispheric portions of the culmen lobule and central lobule, to this investigator's knowledge have not been described, or had any adequate term applied to them. In the GaZago the culmen passes into the hemisphere and decreases in size until be- coming attenuated into a single folium directly adjacent (medially) to the flocculus (Fig. 9). The hemispheric portion of the culmen is roughly pyramidal in shape, the base of the pyramid located at the hemisphere-vermis junction (Fig. 6). The hemispheric extension of the central lobe also comes into close apposition to the flocculus, but it does not taper as it passes laterally (Fig. 5). The vermal portions of the culmen and central lobules are designated as vermal culmen and vermal central lobulesrespectively (Fig. 4, 5). The hemispheric portions of each lobule are designated as culmen pars lateralis and central lobule pars lateralis respectively (Fig. 4, 5). Specific terminology is justified for these regions because it 27 appears that lateral extensions of the vermal culmen and vermal central lobule tend to be primate characteristics since this con— dition is not seen in some advanced subprimate mammalian forms (Larsell, 1953) excepting the monotremes (Dillon, 1962). The culmen, central lobule and lingula have a characteristic number of intrinsic sulci. An intrinsic sulcus is defined as one which does not communicate with the edge of a folium or lobule, or with another fissure or sulcus. In the culmen there are always three (Fig. 5) and occasionally four intrinsic sulci (Fig. 3) dividing the culmen into u—s folia. The most superior sulcus (Fig. 3) is shallow, confined to the vermis, and the one that is usually not present. This sulcus is not illustrated in Figure 5. These sulci extend into the culmen pars lateralis but end before reaching its lateral boundary. The central lobule has two intrinsic sulci, one limited to the vermal central lobule and one continuous into the central lobule pars lateralis (Fig. 5). The lingula has a single intrinsic folium, which is rarely absent (Fig. 3). The fissure between the lingula and central lobule does not continue into the hemisphere, is relatively deep, and distinctly separates the vermal central lobule and the lingula (Fig. 3, 5). Therefore the terms post central fissure and prelingual fissure and considered synonymous. The numerical method of Larsell (1952, 1953) can be applied to the Galqgo with minor modifications. The lingula is composed of sublobules Ia and Ib; the first folium of the vermal central lobule 28 is lobule II and the upper two folia are indicated as sublobules IIIa and IIIb (Fig. 3, S). This division of the central lobule is based on the fact that the first intrinsic sulcus is not only the deepest, but is also continuous into the central lobule pars lateralis (Pig. 5). The vermal culmen is composed, in most cases, of four main folia. Just above the precentral fissure are sub- lobules IVa and IVb, and just rostral to the primary fissure are sublobules Va and Vb (Fig. 3). Sublobule Vb is on rare occasion divided by a very shallow superficial fissure. The same criterion used on the central lobe is applied to the culmen. Those folia separated by sulci that are continuous into the culmen pars lateralis are designated as lobule IV (sublobule a 8 b), and folia separated by sulci limited to the vermal culmen as lobule V (sublobule a 8 b) (Fig. 5). Posterior Lobe A sagittal section of the cerebellum reveals the four main lobules of the posterior half of the Ghlago cerebellum; the median lobule (of Elliot Smith 1903a), the pyramidal lobule, the uvular lobule and the nodular lobule. The pyramidal lobule of the vermis and the nodular portion of the flocculonodular lobule (Crosby at al., 1962) must be considered at this time even though these areas will be discussed again later (of Flocculus and Paraflocculus). The first main division of the posterior lobe is composed of the declive and the tuber (Fig. u, 5). Elliot Smith (1903a) divided the prosimian hemisphere into three areas, merely designating them 29 as areas A, B, and C from rostral to caudal. His area A was the single folium just caudal to the fissure prima, which he later desig- nated as the area lunata (1903b). A more widely used and more accept— able term for this area is lobulus simplex (Fig. u) (Crosby at al., 1962). In the GbZago simplex is represented on sagittal view by the uppermost folium on the caudal wall of the primary fissure. In about half of the Ghlagos examined the lobulus simplex was divided by an intrinsic sulcus into two narrow folia. When this occurs there is a superficial representation of the simplex in the vermis just rostral to the remainder of the declive (Fig. 4). In animals with a single large lobulus simplex there is no superficially visible portion in the vermis. The fissure separating the lobulus simplex from the rest of the hemisphere crosses the midline (Fig. 3) and continues to the lateral extent of the hemisphere (Fig. 4). As this fissure passes from the vermis into the hemisphere it essentially rises out of the fissure prima to become visible on the surface, in doing so it courses slightly caudal, then passes laterally to the margin of the hemisphere (Fig. u, 5). The term posterior superior fissure has been applied to this fissure in the rat, cat and monkey (Larsell 1952, 1953) and it is the post clival fissure in man (Crosby et aZ., 1962). Since there is no anterior superior fissure in these adult animals, the former terminology is confusing. It is therefore sug— gested, in an effort to have adequate yet accurate terminology, that the term simplex fissure be applied to this fissure in the prosimian cerebellum (Fig. 3, u, 5). The main superficially visible region of 30 the vermis caudal to the fissure prima is the remaining portion of the declive (Fig. 3). These two folia have no direct continuation into the hemisphere. Lateral to the declive and separated from the caudal area of the hemisphere by a second deep fissure is the ansi- form lobule (Fig. 9). (It should be noted at this point that the reference to portions of the hemisphere that are "lateral" to portions of the vermis (i.e. declive) is based on the anatomical juxta-position of fissures and folia. This medial-lateral association of one part of the vermis to one part of the hemisphere is not intended to in— dicate a functional association but only a gross anatomical relation— ship based on the continuation of folia, and/or fissures from the vermis onto the hemisphere, or the close apposition of main folia and fissure in these areas. This criterion should be kept in mind during the remainder of all topographical discussion.) In the Galago this lobule is made up of two folia separated by a deep intercrural fissure, the rostral folia is interpreted as crus I and the caudal folia as crus II of the ansiform lobule (Fig. 5). A slight variation of the large single folium making up crus I is noted in the same animals that have a sulcus in the lobus simplex (page 28). In the specimens with a sulcus in the lobulus simplex a very shallow and superficial sulcus is also present in crus I of the ansiform lobule (Fig. n). This is an intrinsic sulcus, and even when it is present the overall size of crus I does not change. Caudal to the declive of the vermis is the single folium of the tuber, also designated as lobule VII. The tuber is occasionally visible on the surface (Fig. 3) but usually it 31 is overlapped by the declive which forces it down into the pre- pyramidal fissure at the sagittal level (Fig. 4, 5). The remaining portion of the hemisphere which is lateral to the tuber and caudal to the ansiform lobule is the paramedian lobule. It is consistently made up of four folia, the last of which is hidden from view within the fissure between the copula and the paramedian lobule (Fig. 5). The deep fissure separating the ansiform and paramedian lobules in the Galago begins in the midline (Fig. 3), passes laterally and caudally at the vermal-hemisphere junction (Fig. 5, 6), and then continues laterally to the margin of the hemisphere (Fig. 5). The term ansoparamedian fissure has been applied to this fissure in the rat (Larsell, 1952), and this term is apprOpriately applied to this fissure in the GaZago (Pig. 4, 5). Elliot Smith (1903a) defined that broad band of cortex connecting the paraflocculus to the pyramidal lobule as the copula pyramidis. In the lesser bushbaby what appears to be the lowest folia of the hemisphere is actually the copula pyramidis (Fig. n). It is slightly foliated by being subdivided by one or two intrinsic longitudinal sulci. The copula will be described in more detail later (cf Flocculus and Paraflocculus). The prepyramidal fissure separates the tuber of the vermis from the pyramidal lobule (Fig. 3, 4). It is moderately deep and has two or three folia within its bounds. Larsell (1952) has shown that in the early embryonic stages of the rat the cerebellum is essentially a bulbus plate crossed by the developing fissures. The prepyramidal fissure appears early and eventually runs from the midline of the 32 developing hemisphere. It separates those embryonic areas which will eventually differentiate into the paramedian lobule and the copula pyramidis. As the paramedian lobule grows it pushes the lateral part of the prepyramidal fissure caudal. The fissure therefore essentially extends from the midline to the hemisphere-vermis junction, continues caudally for a short distance as part of the deep paramedian sulcus (discussed later), then extends laterally to separate the lowest visible paramedian folium from the copula (Fig. H, 5). The term prepyramidal fissure is suggested for and applied to the entire extent of this fissure even to its lateral termination. The pyramidal lobule in Galago is consistently divided into three folia designated as sublobules VIIIa, VIIIb and VIIIc (Fig. 3). These folia are intrinsic to the vermis and, excepting its connection to the copula (discussed later) the pyramidal lobe has no lateral representa- tion in the hemisphere. The sulci of this lobe are also intrinsic and have no lateral counterparts in the hemisphere. On the sloping caudal surface of the Galago cerebellum the hemisphere is sharply separated from the vermis (Fig. u) by a deep groove, the paramedian sulcus. As previously mentioned the prepyramidal fissure assists in the formation of this groove as it courses caudally and laterally (Fig. 13). There is no indication of a paramedian sulcus on the anterior surface of the cerebellum since about half of the anterior lobe (II, IIIa 8 b, IVa 8 b) is directly in contact with the caudal colliculus of the mesencephalon. The postpyramidal fissure separates the pyramid from the uvula 33 (Fig. 3, n). In the white rat embryo the postpyramidal fissure (Fissura secunda of Larsell) extends laterally separating the early paraflocculus into dorsal and ventral parts (Larsell, 1952). In the adult GaZago no lateral extension of the postpyramidal fissure is seen. In both macroscopic observations and in tracing this fissure from medial to lateral in microscopic serial sections it appears to stop at the vermis hemisphere junction. Two possible explanations of this are offered later (of Flocculus and Paraflocculus). It is possible that the lateral portions of the postpyramidal fissure are represented by the groove between the copula pyramidis and the flocculonodular bundle. The uvula has three folia designated as sublobule IXa, IXb and IXc, which do not have hemispheric representation in the adult GaZago. The nodulus is separated from the uvula by the prevodular fissure, one of the deepest fissures in the adult cerebellum (Fig. 3, 5). The term posterolateral fissure has been applied to that fissure separating the uvula-ventral paraflocculus portion of the developing cerebellum from the nodulus-flocculus portion (Larsell, 1952). In the adult GaZago this fissure does not appear to continue into the hemisphere, but rather it ends just lateral to the nodulus. The details of the flocculonodular relationships are discussed under the heading Flocculus and Paraflocculus. The most caudal lobule of the cerebellar vermis is the nodulus (Fig. 3). It invaginates the posterior medullary velum and helps to form part of the caudal wall of the fourth ventricle. The nodulus 31+ is designated as lobule X with no sublobules being indicated since these subdivisions were not entirely consistent. There were usually two subdivisions, a large caudal sublobule and a larger rostral sub- lobule, sometimes three sublobules were seen and rarely four. In view of the inconsistent pattern, the only designation of the nodulus will be lobule X. ANS CVL CLPL CP CPL Cr.I Cr.II CV PAP FP FPccn. FPn. FPo. FPoC. FPr. PS LS PM PMS PY II III IV VI VII VIII IX ABBREVIATIONS FOR FIGURES 8, 4 and 5 ans iform lobe vermal central lobule central lobule pars lateralis c0pula pyramidis culmen pars lateralis crus I of ansiform lobule crus II of ansiform lobule vermal culmen declive ansoparamedian fissure fissura prima precentral fissure prenodular fissure postpyramidal fissure postcentral fissure prepyramidal fissure simplex fissure lingula lobus simplex nodulus position of the parafloccular stalk paramedian lobule paramedian sulcus pyramidal lobule tuber uvula lingula: sublobules Ia, Ib ventral central lobule dorsal central lobule: sublobules IIIa, IIIb ventral culmen: sublobules IVa, IVb dorsal culmen: sublobules Va, Vb declive: sublobules VIa, VIb tuber pyramidal lobule: sublobules VIIIa, VIIIb, VIIIc uvula: sublobules IXa, IXb, IXc nodulus 35 CV’ b a b V ¥ , IV FPcN 5 II III CIV' a II I . b c'/ FPoC \\\‘ u. ‘ly,,/’*\\\\\\ FS LS D c VI PAP T VII FPr a b Py VIII c w4-—FPo 0 IX b U c / Figure 3. Mid-sagittal view of the cerebellum of the lesser bushbaby (Gblago). 36 Figure u. Lateral view of the cerebellum of the lesser bush— baby (Galago). The paraflocculus has been removed (P) for sake of clarity. The inset illustrates the variation seen in the lobus simplex (LS) and crus I of the ansiform lobule. It is noted that when the lobulus simplex is fissured it is also superficially visible in the vermis. 37 ‘— FPo FP FAP / /'l ANS CV rs FP / 4—- FPcn ClV <— FPoC Figure 5. A semi-diagrammatic drawing of the cerebellar cortex in the lesser bushbaby (Gulago). This drawing represents only those portions of the cortex that are superficially visible, and it is longitudinally flattened so the entire cortex is illustrated on one plane. 38 TUPAIA GLIS AND TUPAIA CHINENSIS The tree shrews used in this study are adult; therefore all morphologic points of discussion, both gross and microscopic are for the adult forms. Any attempt to suggest the embryological develop- ment of a particular part is pure conjecture. The discussion of the gross anatomy of the Tupaia cerebellum includes T. glis and T. chinensis, since it is usually applicable to both species. When specific and consistent differences are noted between these animals they are discussed separately and a conclusion is drawn concerning a particular point of difference. In the overall plan the cerebelli of these two Species of Tupaia are strikingly similar, and only a few differences are noted. The tree shrew (T. gZis and T. chinensis) is found in India, throughout South East Asia, Thailand and on many of the off—shore islands in this region (Napier and Napier, 1967). The tree shrew is about l/H smaller than the Galago, quadrupedal, and moves about with "jerky scurrying movements" such as a rodent. Zapata may jump from branch to branch over short distances but is believed to move about mainly on the forest floor (Napier and Napier, 1967). In the tree shrew, because of its mode of locomotion, there is less need for balance and fine digital sense, a point of particular significance. flid-sagittal section A mid-sagittal view of the Tupaia cerebellum reveals a moderately complex structure compared to the GaZago and rat (Larsell, 1952) (Fig. 6). A distinct deep primary fissure is present and it appears 39 90 to divide the cerebellum into almost equal anterior and posterior lobes. However this fissure does not pass into the hemisphere but extends rostrally over the anterior surface of the hemisphere ending at what essentially is the junction of the hemisphere and vermis (Fig. 8, 9). Consequently the sulci within the fissure prima are restricted to the vermis. Anterior Lobe L- The anterior lobe of the tree shrew cerebellum is divided into a culmen, central lobule and lingula by the precentral and postcentral fissures respectively (Fig. 6, 7, 9). The criterion for this method of naming fissures is discussed on page 23. Considering the position of the primary fissure it is therefore noted that the anterior lobe is restricted to the vermis (Fig. 8, 9). The lingula invaginates the anterior medullary velum and is distinctly separated from the central lobule by a deep prelingual (postcentral) fissure (Fig. 6). The lingula has no intrinsic sulci and no continuations into the hemisphere (Fig. 8). The method of Larsell (1952) is applicable to the tree shrew when slightly modified, thus the lingula is indicated as lobule I (Fig. 6). The central lobule is completely separated from the culmen and lingula by the precentral and postcentral fissures respectively (Fig. 6, 9). A characteristic number of intrinsic and extrinsic fissures are noted for the central lobule. The term, intrinsic sulci, has been defined, and an extrinsic sulci is defined as one which reaches the edge of a main lobe (or lobule) thus dividing the 41 lobe (or lobule) in a more definite and distinct manner. The central lobule has a single extrinsic sulcus dividing it into a large upper portion and a slightly smaller lower portion (FTC—Fig. 7). The upper and lower portions of the central lobule are each divided by a single intrinsic sulcus (Fig. 8, 9). The intrinsic sulcus of the upper por— tion will sometimes reach the margin, however when it does it is very shallow at this point, and even this very shallow continuation is sometimes absent. The entire central lobule of Tupaia is vermal (i.e. vermal central lobule) and there is no hemispheric representa- tion (i.e. central lobule pars lateralis). The lower portion of the vermal central lobule consists of two folia designated as sublobules 11a and 11b (Fig. 6). The upper portion of the vermal central lobule, even though collectively it is larger, also consists of two folia designated as sublobules IIIa and IIIb (Fig. 6). The intercentral fissure is the division between the upper and lower portions of the vermal central lobule because it is not only the deepest fissure but it is also the only fissure to consistently reach the lateral margin of the central lobule. The culmen (vermal culmen) is separated from the posterior half of the cerebellum by the deep primary fissure and from the central lobule by the moderately deep precentral fissure (Fig. 6, 8, 9). The vermal culmen of Tupaia is also characterized by a single distinct extrinsic fissure dividing it into about equal halves. This inter— culmen fissure reaches the lateral extent of the culmen dividing it into an upper and lower portion (Fig. 9). These two regions end 42 before becoming part of the hemisphere. The ventral portion ends before reaching the margin, while the dorsal portion extends later- ally then anteriorly to 7the margin of the cerebellar cortex (Fig. 9). In one brain of Tupaia chinensis both portions of the culmen ex— tended (anteriorly) to the margin of the cerebellum, so a small degree of variation can be expected to occur at this point. The upper and lower portions of the vermal culmen are each divided by a single intrinsic sulcus. These sulci travel only a short distance from the midline, then disappear (Fig. 7c8). They are consistent and indicated as sublobules IVa and IVb for the lower portion and sublobules Va and Vb for the upper portion of the vermal culmen (Fig. 6). The primary fissure separates what has been shown to be a rather modest anterior lobe from the expanded hemisphere. The anterior lobe is considered to be modest in its development because it essentially has no more than the vermal components of all its lobules, with no representation in the hemisphere. Posterior Lobe The posterior lobe of the cerebellum of the tree shrew (excepting the technicality of a flocculonodular lobule) is composed of a median lobule (declive-tuber), a pyramidal lobule, a uvular lobule and a nodular lobule (Fig. 6,7,8). Since there is no culmen pars lateralis and central lobule pars lateralis, the primary fissure extends from the midline laterally then rostrally (essentially over the front of the hemisphere) separating the hemisphere from the vermal culmen #3 (Fig. 7,8,9). The primary fissure is deep on the sagittal plan but rapidly diminishes in depth as it progresses laterally. Directly caudal to the fissura prima on the vermis are the flat folia com— prising the lobus simplex of the declive (Fig. 7,8). In the Tupaid glis the lobus simplex portion of the declive is composed of two folia. The sulcus separating these two folia is shallow and is not continuous with its counterpart from the hemisphere (Fig. 7). In It gZis these two medial folia of the lobulus simplex fade into a single large lateral lobulus simplex which extends laterally and rostrally onto the anterior surface of the hemisphere (Fig. 8,9). In Tupaia chinensis the lobulus simplex of the declive is a single broad folium that is as wide as the lobulus simplex of T. gZis, however it is not divided by the shallow intrinsic sulcus (Fig. 6). The simplex sulcus (Fig. 8C9) begins at the hemisphere-vermis junction and extends from the dorsal surface of the hemisphere (Fig. 8) onto its anterior surface (Fig. 9). The term lobulus simplex is applicable to both the vermal and hemispheric portions of this region. Based on the com- parison of the lobus simplex of T. gZis with T. chinensis, this region is designated as lobule VIa in Tupaia chinensis and in Tupaia glis. The reason for this will soon be apparent. Caudal to the superficially visible lobulus simplex is the re— mainder of the declive region of the vermis. It is made up of one Slightly larger folia, and two slightly smaller ones (Fig. 7,8). These are designated as sublobules VIb, IVc and IVd (Fig. 6,8). Sub- lobule VIb is occasionally subdivided by a very shallow snicus, 41+ however this does not appear to be a significant occurrence, and none of the sulci of the declive are directly continuous into any corresponding fissure or sulcus in the hemisphere (Fig. 8). The single folium caudal to the declive of the vermis and slightly enlarged in its lateral extent is the tuber (Fig. 7, 8). It is undivided and is usually separated from the hemisphere by a consistent shallow sulcus (Fig. 8). The tuber is separated from the pyramidal lobule by the deep prepyramidal fissure and being consistently single is designated as lobule VII. The reasons for suggesting the two folia of the lobus simplex of T. gZis as sublobule VIa are now apparent. In both T. gZis and T. chinensis the declive is composed of three folia and the tuber of one. Even though the lobulus simplex of T. chinensis is not sub— divided as it is in T. gZis it is just as large. Secondly the subdivision in T. gZis, when viewed by itself, does not appear to be a significant sulcus. It is therefore concluded that this division of the vermal portion of the lobus simplex in T. gZis is only a slight modification of a distinct general plan and not worthy of the classification of sublobules. Even though none of the sulci of the vermis are directly con- tinuous with those of the hemisphere, there are two main fissures which come into close apposition with the vermis (Fig. 8). The first is the simplex fissure, the second is the ansoparamedian fissure (Fig. 7, 8). By nature of their configuration and apposition to the fissures and sulci of the vermis, it appears that the declive of the 45 vermis is adjacent laterally partially with the ansiform lobe and partially with the paramedian lobe (Fig. 7, 8). The ansiform lobule of the Tupaia is made up of five folia (Fig. 7, 8). This lobule is distinctly separated from the rest of the hemisphere by the simplex fissure and ansoparamedian fissure. Within this large ansiform area, a fissure of secondary importance is noted. It is deep, approaches more closely to the vermis than any other fissure within the lobe, and consistently divides the rostral three ansiform folia from the caudal two folia (Fig. 7, 8). For clarity sake this cleft will be termed the intercrural sulcus (after Larsell, 1952, 1953). The intercrural sulcus divides the rostral three folia, interpreted as crus I, from the caudal two folia interpreted as crus II (Fig. 7). In both of the Tupaia in this study crus I is consistently composed of three folia, of which the middle one is the largest. Crus II in Tupaia glis is made up of two folia separated by a shallow sulcus (Fig. 7) while in Tupaia chinensis the sulcus is absent but the resulting single folium is wide. Crus I and crus II of the ansiform lobe make up the majority of the anterior and lateral surface of the hemisphere indicating a state of relatively advanced develOpment for this particular region. The large portion of the hemisphere lateral to the tuber and caudal to the ans0paramedian fissure is the paramedian lobule (Fig. 7, 8). The paramedian lobule is made up of four caudally buldging folia making UP the caudal surface of the hemisphere, and separated from the vermis by a deep groove, the paramedian fissure (Fig. 7, 8). The uppermost |+6 folium of the paramedian lobule extends above the level of the ansiform lobes crus I and crus II (Fig. 9). There is no indication of a para— median groove, sulcus, or fissure on the dorsal or anterior aspects of the Tupaia cerebellum (Fig. 7,9). The prepyramidal fissure separates the tuber from the pyramidal lobule of the vermis. This fissure is moderately deep on the midline, extends laterally then caudally (as part of the paramedian fissure) ' then again courses laterally around the hemisphere separating the paramedian lobule from the copula pyramidis (Fig. 7,8). It is suggested that the term prepyramidal fissure be used for this fissure from the midline to its lateral extent. The pyramidal lobule in T. gZis and T. chinensis is consistently composed of two folia of about equal size. These are designated as sublobules VIIIa and VIIIb of the pyramidal lobule. What appears to be the lower most folium of the hemisphere is part of the copula pyramidis (Fig. 7). In the tree shrew the copula is differentiated into a copula pyramidis lateralis and a copula pyramidis medilis (Fig. 7). The copula pyramidis lateralis is a small tear—drop shaped folium very closely associated with the para- flocculus and overhung by certain regions of the ansiform lobule. The copula pyramidis medialis is the narrow band of cerebellar cortex that is continuous with the pyramidal lobule of the vermis. The details of these relationships will be discussed later (of Flocculus and Paraflocculus). The postpyramidal fissure separates the pyramidal lobule and #7 the uvular lobule, and in the adult animal has no macroscopically lateral continuation. The uvula of Tupaia gZis is usually composed of what appears to be four main folia (Fig. 6). The uvula of Tupaia chinensis is made up of only two main folia and appears, with the exception of the extra folia, to closely resemble the same structure in T. glis. (Fig. 6). Based on this comparison within Tupaia this sulcus is termed the interuvular sulcus, the sublobules are desig— nated IXa and IXb (Fig. 6). It should again be noted and emphasized that the variation between the uvula of T. glis and that of T. chinensis is relatively minor while the general concept of two sublobules is quite acceptable. The two extra intrinsic sulci in the uvula of T. gZis are not deep significant sulci therefore the added sublobules created by these are not considered significant enough to merit a separate classification. The prenodular fissure also appears to have no lateral con— tinuation yet affords a distinct separation between the uvula and nodular lobules. The nodulus of the tree shrew appears to have a fairly consistent pattern. In T. gZis two large ventral folia and an occasional small dorsal folium is noted. The T. chinensis has only the two ventral folia. In view of this consistency of a general pattern these folia are designated as sublobules Xa and Xb. The sulci of the uvula and nodulus are intrinsic and these lobules have no representation in the hemisphere. The flocculo—nodular relation— ships are discussed later (of Flocculus and Paraflocculus). ABBREVIATIONS FOR FIGURES 6, 7, 8, and 9 ANS ansiform lobule CLV - vermal central lobule CP - copula pyramidis CPL - copula pyramidis lateralis CPM - copula pyramidis medialis Cr. I - crus I of ansiform lobule Cr. II - crus II of ansiform lobule CV - vermal culmen D - declive FAP - ansoparamedian fissure FIC - intercentral fissure FICL - interculmen fissure PP - primary fissure FPcn. - precentral fissure FPn. - prenodular fissure FPo. - postpyramidal fissure FPoC. - postcentral fissure FPr. - prepyramidal fissure FS - simplex fissure' ICS - intercrural sulcus IUS - interuvular sulcus L - lingula LS - lobus simplex N - nodulus P - position of the parafloccular stalk PM — paramedian lobule PMS - paramedian sulcus T -tmmr U — uvula I - lingula II - ventral central lobule: sublobules 11a, 11b III - dorsal central lobule: sublobules IIIa, IIIb IV - ventral culmen: sublobules IVa, IVb V - dorsal culmen: sublobules Va, Vb VI - lobulus simplex and declive: sublobules VIa, VIb, VIc, VId VII - tuber VIII - pyramidal lobule: sublobules VIIIa, VIIIb IX - uvula: sublobules IXa, IAb X - nodulus: sublobules Xa, Xb 48 FP _, vr’a O D b C d VI‘ T VII 0 P b VIII 0 /-\ / IX " , b II a X Figure 6. Mid-sagittal view of the cerebellum of the tree shrew (Tupaia gZis). The upper inset shows the variation of the lobus Simplex seen in T. chinensis. The lower inset shows the variation of the uvula and nodulus also seen in T. chinensis. See text for discussion. ‘ 1+9 Lateral view of the cerebellum of the tree shrew (Tupaia glis). The paraflocculus (P) has been Figure 7. removed for sake of clarity. Ansiform lobule crus II is partially divided by a shallow sulcus in T. glis, however it is usually a single wide folium in T. chinensis. 50 A semi-diagrammatic drawing of the cerebellar Note particularly This drawing Figure 8. cortex in the tree shrew (Tupaia glis). the anterior lobe, and the ansiform lobule. represents those portions of the cortex that are super- ficially visible, and it is longitudinally flattened so the entire cortex is illustrated on one plane. 51 K25 ,1 g ’ .t 7}? ANS-Cr I 'I CLV’ Figure 9. Anterior view of half of the cerebellum of the tree shrew (Tupaia glis). Note the relatively large size of the paraflocculus. 52 CORRELATIVE DISCUSSION When contemplating the morphology of the prosimisn cerebellum one is particularly aware of the fact that animals not closely related to primates on the phylogenetic scale, such as cats and dogs (Reighard and Jennings, 1940; Miller et aZ., 196u) and especially the cetacean (Jansen, 1950), possess a cerebellum of rather complex development. On the other hand many of the lower primates have a rather simple cerebellum. To amply explain the differences would in- volve a study of the olive, vestibular nuclei and pontine nuclei and spinocerebellar tracts, which is beyond the limits of the present study. Elliot Smith (1903a) was of the opinion that the development of the cerebellum depended on the activities of the animal as well as its zoological rank. In a series of classical stimulation and lesion studies of the cerebellum, Mussen (1967) attributed general functions to specific regions. The pyramidal lobule functioned in the proper maintenance of backward balance, while the culmen and central lobules are respon— sible for forward balance. A lesion of the pyramidal lobule would cause the animal to fall backward when it attempts to assume an erect posture. A lesion of the culmen and central lobule would result in the animal running into things or falling forward. Brodal (1967) has shown that the dorsal and ventral spinocerebellar tracts project almost exclusively to the culmen, central lobule and lingula, the pyramidal lobular and to the paramedian lobule to a limited extent. Oscarsson (1965) has shown that the dorsal spinocerebellar tract as well as the 53 51+ cuneocerebellar tract terminate in the intermediate region of the ipsilateral anterior lobe, with some terminations in the adjacent cortex of the vermis. According to Mussen (1967) using a stimulation technique on the paramedian lobule these ascending fibers are specif- ically localized to certain lobules. "The dorsal lobule is concerned with movements of the shoulder, the middle lobule with the muscu— lature of the upper leg and the ventral lobule with the activities of the foreleg and paw." The irregular movements in these lesions studies were seen in the cat, and seen ipsilateral to the lesion (Mussen, 1967). In the overall comparison of the gross topography of the GaZago and Tupaia cerebelli (exclusive of the flocculus and paraflocculus) several significant differences are noted. The anterior lobe of Tupaia is entirely vermal in its location while the same area in the GaZago has relatively large lateral extensions. The pyramidal lobule of Tupaia consistently has two folia and the pyramidal lobule of Galago has three folia. For the Galago this is interpreted as advanced development not only because of the extra folium, but because the pyramidal lobule occupies considerably more of the posterior lobe in Galago than in Tupaia. If the work of Mussen (1967) on the cat can be extrapolated to the prosimian, the following explanation is offered. The anterior lobe of the cerebellum and the pyramidal lobe are responsible for forward and backward balance respectively. In its mode of locomotion the Galago needs a keen sense of balance, whereas the Tupaia has little need for such a keen sense since it is quadra— pedal. In the rat, which progresses much as the Tupaia, there are 55 lateral portions of the central and culmen lobules, however in this animal the vermal portions of these lobules are not as highly differ- entiated as in the Tupaia (Larsell, 1952; Zeman and Innes, 1963). Furthermore the pyramidal lobule of the rat is composed of one folium, while in Tupaia there are two, and in Galago three. In a study of fourteen different groups of insectivores representing both "basal" and "higher" forms (Stephan and Andy, l96u) Le Gros Clark (1932) noted that either the pyramidal lobule is a single folium, or it is not even differentiated (by a dividing fissure or sulcus) from the other more rostral portions of the posterior lobe. In these forms the anterior lobe is also poorly differentiated. The advanced development of the anterior lobe and pyramidal lobule of the GaZago over the Tupaia is interpreted as primary traits (anatomy) resulting from a demand upon the animal (i.e. Galago) by its environment. This is a type of adaptive radiation (Buettner-Janusch, 1966) in which a primary trait (those portions of the cerebellum predominately related to locomotion) undergoes modification in order for the animal to adequately adapt to a changing environment. In Tupaia these adaptive changes apparently did not take place. Therefore this animal is rele- gated to a state of quadrapedal locomotion with an inability to assume a stable erect posture or progress with great space encompassing leaps as does the Gulago. To further corroborate this interpretation of the anterior lobe in Galago and Tupaia, the same lobes in Tursius are noted. The Tarsius inhabits arboreal regions of "secondary growth" and pro- gresses by strong, powerful leaps in both horizontal and vertical 56 directions (Hill, 1955). According to Woollard (1925) the hemispheric portions of the culmen and central lobule extend laterally and come into close apposition to the flocculus, very similar to the condition seen in the Galago. From this comparison it can be seen that in prosimians the development of the anterior lobe is possibly, in part, related to locomotion. A second region deserving a special morphological comparison is the ansiform area crus I and crus II. In the GaZago the ansiform lobule is composed of only two folia, while in T. chinensis it is made up of four folia and in T. gZis of five. In the Tupaia of this study the entire hemisphere rostral to the ansoparamedian fissure is the ansiform lobule. The same region of the hemisphere in Galago in- cludes the central lobule pars lateralis, culmen pars lateralis, and ansiform lobule. It has been previously suggested that the expanded anterior lobe in Galago is directly related to locomotor pattern; now an explanation is fought for the large ansiform lobule in Tupaia. The ansiform lobule in higher mammals, especially higher primates, and man is greatly expanded taking up a large percentage of the cerebellar hemisphere (Crosby et al., 1962). In the cat (Brodal, 1940a) it has been shown that the dentate nucleus (lateral nucleus of Flood and Jansen, l961) projects to the entire ansiform lobule. It has also been shown (Brodal, 19u0b) that the ventral lamina of the principle olivary nucleus projects to the ansiform lobule crus II and the dorsal lamina projects to crus I. A second region projecting to the ansiform lobule is the area of pontine nuclei. In the cat and rabbit these 57 projections pass to the entire ansiform lobule and to most of the 'paramedian lobule (Brodal and Jansen, 1996). A third nuclear region of the brain stem projecting directly to the cerebellum is the vesti- bular nuclear masses. However fibers from the vestibular nuclei project to the flocculus, paraflocculus, nodulus, uvula and fastigii nucleus (Brodal and Hoivik, 196“; Brodal, 1967; Carpenter, 1967). It has also been previously noted that there is a direct relationship between the deve10pment of the flocculus and the vestibular mechanism, and between the flocculus and the coordination of eye movement. The morphological significance of this will be discussed later (cf Flocculus and Paraflocculus. In the above discussion it has been shown that there is a correlation between the development of the brain stem nuclei and the related areas of the cerebellar cortex (e.g. vestibular, olive, etc.). When considering the evolution of the cortico-ponto and ponto-cerebellar system, the gross development of the pons is a relative indicator of the complexity of this association pathway. Marsden and Rowland (1965) compared the gross development of the pons, olive, and pyramid over a wide range of mammals including the Loris, Lemur and greater Galago. According to Marsden and Rowland (1965) in most sub—primate forms the pons and olive are poorly de- velOped; the olive barely visible and the pons usually pretrigeminal. These authors noticed that in primates, beginning with the Lemur, there is a rapid and progressive increase in the size of the pons and olive. In the Lemur the pons is 1/4 post-trigeminal, and in adult men it is 2/3 post-trigeminal. The reader should note that the Lemurs 58 are phylogenetically close to Tupaia. To adequately answer the question of the advanced differentiation of the ansiform lobule, a study is needed of the pons, olive, and particularly the pontine nuclei. This is considerably beyond the scope of this present study. For the present it is sufficient to point out that those regions of the cerebellar cortex associated with the pontine and olivary nuclei show an advanced degree of development in the Tupaia. This implies a reasonable degree of differentiation of the respective brain stem regions, and the possible existence of an advanced cortico-ponto- cerebellar pathway. Larsell (1952, 1953) studying the rat and rhesus, and Larsell and Jansen (1967) the avian noticed the relationship between the pubococcygeal muscles and the anterior lobe of the cerebellum, par- ticularly the lingula. Chang and Reich (1949) demonstrated strong projections from the caudal segments of the cord to the lingula in the spider monkey, an animal with a prehensile tail. In Tupaia there is a single folium while GaZago has two folia in the lingula. The tree shrew, because of its quadrapedal mode of locomotion, does not use its tail as extensively for balance as does the GhZago. The bushbaby in its leaping about and sitting uses its tail extensively for balance. It is the author's Opinion that the difference in differentiation of the lingula has been partially governed by the evolutionary development of the locomotor pattern of each animal. In the Gblago, sometimes the lobulus simplex is not superficially visible in the vermis and consequently was not given classification 59 as a sublobule. However at this point, for clarity it will be desig- nated as sublobules VIa with the remaining portions of the declive being sublobules VIb and VIc. This follows the general method of Larsell (1952). Regardless of sublobule classification, the simplex lobule should be called lobulus simplex both in its hemispheric and vermal region (Crosby et aZ., 1962). The following listing shows the relative comparative complexity of topographical development in Ghlago and Tupaia. Galago Tupaia Ia, Ib lingula I II ventral central 1. Ila, IIb IIIa, IIIb dorsal central 1. IIIa, IIIb . IVa, IVb ventral culmen IVa, IVb Va, Vb dorsal culmen Va, Vb VIa, VIb, VIc simplex 8 declive VIa, VIb, VIc, VId VII tuber VII VIIIa, VIIIb, VIIIc pyramidal VIIIa, VIIIb IXa, IXb, IXc uvula IXa, IXb X nodulus Xa, Xb Oddly enough this comparison shows twenty distinct and consistent sublobules for each animal on a mid-sagittal section. A diagrammatic comparison of the cerebellar hemispheres of Tupaia and Galago follows. Galago Tupaia CentraI lobules pars lateralis Culmen pars lateralis Crus I Ansiform CPUS I AnSIform Crus II Crus II Paramedian lobule Paramedian lobule 60 Two anatomical criteria are suggested as evidence for the approximation of Tupaia to the primates; [l] the overall complexity of topographical differentiation, and [2] the advanced differentiation of the ansiform lobule. It is this investigator's opinion that these two factors indicate primate tendencies with a rather distinct advance- ment over the Insectivores. This conclusion is also drawn when com— paring GaZago and Tupaia. Perhaps this is a step toward answering the question of Hoffman (196u) concerning increased deveIOpment of the prosimian cortico—ponto-cerebellar system. "With reference to the in— creased size (quantitative) of the cerebellum: Is this increase in the cerebellum particularly associated with an increase in the cortico- ponto-cerebellar system with the largest increase in the cerebellar hemispheres and systems through the superior cerebellar peduncle?" The problem with the term secondary fissure has been discussed. Shortly after its introduction in prosimians by Elliot Smith (1903a) Bradley (190“) suggested that the term had no inherent significance. Larsell (1952, 1953) pointed out that the fissura secunda is the fourth to appear embryologically and therefore the term does not in— dicate sequence, nor is this fissure necessarily the second deepest. Furthermore there is disagreement in the literature on what fissure is the "fissura secunda". It is therefore strongly suggested that this term be replaced by the term postpyramidal fissure as previously pr0posed in this study. The latter term has merit if for no other reason than its descriptive quality. FLOCCULUS AND PARAFLOCQULUS GENERAL INTRODUCTORY REMARKS The flocculus and paraflocculus are phylogenetically some of the oldest parts of the cerebellum, even though in some lower forms they are almost grossly imperceptible (Nieuwenhuys, 1967). In the structurally simple mammalian cerebellum (rat - Larsell, 1952) the flocculus develOps from a band of cortex directly associated medially with the nodulus of the vermis, and the paraflocculus de- velOpes from the cortex that is medially associated with the uvula (ventral paraflocculus) and pyramid (dorsal paraflocculus). In the adult animals used in this study no distinguishable dorsal and ventral paraflocculi are noticed. Consequently there is no lateral continu- ation of the postpyramidal and prenodular fissure. There are two possible explanations for no lateral extension of the postpyramidal fissure. First, it could be so poorly developed in the embryo that it readily disappears in the adult, or secondly, since there is no grossly distinguishable dorsal and ventral paraflocculus in the adult this region could have a different sequence of embryological develop- ment than the analogous region in the rat (Larsell, 1952). Stroud in 1895 designated the terms flocculus and paraflocculus, whereas two years later Ziehen referred to these collectively as the floccular lobe (Bradley, 1904). Elliot Smith (1902a, 1903a) also applied the term "lobus flocculi" to both of these regions. Bradley (1904)however, stressed the importance of separate terminology and advocated the use of the terms flocculus and paraflocculus. 61 62 Subsequent investigators use this separate terminology. The flocculus and paraflocculus of the GhZago and Tupaia are considered separately from the rest of the cortex so that they can be discussed in detail with a minimal amount of confusion. It must always be kept in mind that these regions are an integrated part of the cerebellar cortex. Because of its intimate anatomical relation- ship the copula pyramidis forms an important part of the following discussion. GALAGD SEWEGMLEWSIS The flocculus and paraflocculus are the most lateral portions of the cerebellum and the paraflocculus is encased in bone. The flocculonodular portion of this complex composes the archicerebellum, or the vestibulocerebellum (Brodal, 1967). Flocculus The flocculus of the GhZago is made up of four folia (Fig. u - p. 36, 10) three of which are incompletely separated from each other. The uppermost folium bulges rostrally, becomes constricted under the stalk of the paraflocculus, and fades out as a narrow band of grey matter passing under the c0pula pyramidis (Fig. u - p. 37). This upper folium is always separated from the remainder of the flocculus by a shallow, yet consistent sulcus. The other two sulci dividing the remaining three folia are essentially intrinsic (Fig. 10). There is a certain amount of variability in the gross structure of the flocculus. To exPlain this as succinctly as possible the folia of the flocculus are designated fl, f2, f3, and fk from upper rostral to caudal 63 (Fig. 10, ll). Folia fl and f2 vary in size, relative to each other, with f2 usually being the larger. Folium f4 occasionally is merely a narrow slip of cortex, subsequently folium f3 is always the largest of the flocculus (Fig. 9 - p. 37, 10). The flocculus is separated from the paraflocculus by a deep non-descript groove labeled as the floccular fissure by Le Gros Clark (1932). This cleft is acknowledged in the Galago (Fig. 10, 11), however the reason for a specific termin- ology for this cleft is questionable. The flocculus is directly applied to the lateral surface of the brain stem (and cerebellum), while the paraflocculus is located in an osseous fossa. Therefore the bony ridge of the osseous parafloccular fossa is a decisive and natural boundary. Paraflocculus The paraflocculus of the Galago is affixed to the lateral side of the cerebellum by a medullated stalk (Fig. A — p. 37, 12). The folia making up the paraflocculus collectively form a slight medially concave structure (Fig. 12). There are five main folia composing the paraflocculus and these are designated pl, p2, p3, p4, and p5 from rostral to caudal (Fig. 10, 12). Folium p1 is occasionally sub- divided by a shallow incomplete sulcus into a larger medial portion and a smaller lateral portion. This is interpreted as a variation in folium pl based on the location and extent of this intrinsic sulcus (Fig. 10, ll, 12). The sulci between the remaining folia are so deep that the medullated substance of the paraflocculus is visible when these folia are separated. When viewing the paraflocculus from its 6H dorsal and ventral surfaces the characteristic number of folia are noticed. An additional folium closely associated with the para— floccular stalk is seen on the ventral view (Fig. 12-B). This is a small, low folium that extends onto the parafloccular stalk, and has no dorsal counterpart. Folium pH projects caudally in a rather distinct manner. No dorsal and/or ventral paraflocculus can be distinguished in the adult Galago even though these probably occur in the embryological state. A groove is present on the ventral surface of the paraflocculus, and from its position the dorsal and ventral contributions to the paraflocculus can be suggested. If the parafloccular stalk were divided in the middle and the division ex- tended throughout the course of the parafloccular groove (dotted line Fig. 12-B) the paraflocculus would be divided into three rostral (upper) folia and two ventral (lower) folia. The three upper folia (dorsal paraflocculus) are pl, p2, and p3, and the lower folia are pu, p5 and the stalk folium. It must be emphasized that this sug- gestion is based solely on the adult, and confirmation or dismissal of this opinion must await the study of embryological material. The deep cleft.separating the paraflocculus from the lateral hemisphere has been termed the parafloccular fissure by Le Gros Clark (1932) (Fig. 11). It should be noticed that this groove is filled by the dorsal edge of the osseous fossa of the paraflocculus. The paraflocculus is attached to the cerebellar hemisphere by a short stalk of fibers. Directly surrounding the stalk is a non— descript region of grey matter (Fig. 9 - p. 37). This low region of 65 cortex is very limited in its extent in most cases. Adjoining this limited region of grey matter is an elevated ridge coming from the copula pyramidis (Fig. 4 - p. 37, l2-A). (This ridge is indicated by a small r in all figures.) By way of this distinct ridge from the copula onto the stalk of the paraflocculus the continuity of these two structures is obvious. In the Ghlago the cepula extends further laterally than any other part of the cerebellar hemisphere. As the cortex of the copula beComes the ridge joining the circum- peduncular grey, it proceeds rostrally and turns slightly medially (Fig. u - p. 37, l2-A). This lateral appearance of the copular ridge and its continuation with the parafloccular stalk is character- istic for the GbZago. Cgpula_Pyramidis It has been previously stated that what appears to be the lower most folia of the hemisphere in Galago, is in fact the copula pyramidis. (For a definition of that portion designated as copula see p. 10.) A caudal and slightly ventral view of the left hemisphere shows the characteristic appearance of the paraflocculus, copula, paramedian lobe and the last three main portions of the vermis (Fig. 13). The copula is distinctly visible from the lateral copular ridge until the cortex extends into the prominent paramedian groove (or sulcus) (Fig. 19). Portions of the paramedian lobule were removed and the pyramidal, uvular, and nodular lobule of the vermis retracted to show the entire lateral-to—medial relations of the copula (Fig. 1A). The copula has usually one or occasionally two intrinsic sulci which run 66 horizontal to its long axis. The intrinsic sulci of the pyramidal lobule stop within the gross limits of this lobule and are not con- tinuous with sulci of the copula. The continuation of the copula into the pyramidal lobule is relatively wide and flat with medullary substance clearly visible at the junction (Fig. 14). Elliot Smith (1903a), when he originally described the c0pula pyramidis, con- sidered the structure significant, however later concluded that it had little significance (Elliot Smith, 1903b). ’This study reveals that perhaps his earlier opinions were more accurate. Kanagasun- theram and Mahran (1960), in a study of the gross anatomy of the nervous system of the lesser bushbaby, erroneously stated "...the c0pula pyramidis which is hidden from the surface by the lateral portions of the middle lobe." Krishnamurti (1966), in a study of the brain of the slow loris (Nycticebus coucang), stated "...the copula pyramidis, a part of which could be seen on either side of the vermis and the rest of it is hidden from the surface by the lateral portions of the middle lobe." As pointed out above, the copula in the Galago is not hidden from view, but is a prominent superficial landmark, therefore the present study is not in agreement with these previous investigators. The copula pyramidis of the GaZago is usually made up of two folia although occasionally three were present (Fig. 4 - p. 37). Since the prepyramidal fissure is deep, the paramedian lobule appears to be resting on the c0pula. The intrinsic sulci of the copula extend the entire width of the structure in about half of the Specimens. When this is the case, the copula made up what 67 appears to be the lower most two folia of the hemisphere. The extent of the intrinsic sulci of the copula is somewhat variable and this region merits careful examination before its boundaries are accurately determined, otherwise the copula cOuld be mistaken for lower portions of the paramedian lobule. The fiber connections of the flocculus and paraflocculus of the Galago can-best be determined definitively by the utilization of a degeneration technique. In the present study it has been noted that the postpyramidal and prenodular fissures have no lateral extensions. In a series of dissections of several cerebelli, which involved re- moval of the copula, a direct communication between the flocculus— paraflocculus and uvula-nodule is seen (Fig. 15). The postpyramidal fissure appears to stop since there is not a detectable fissure be- tween the copula and underlying bundle of fibers. If one wishes to consider the anatomical apposition of lthe copula and the fibers of the flocculonodular bundle as the lateral representation of the postpyramidal fissure, this is indicated in figure 15. The band of fibers connecting the flocculus and paraflocculus with their respec— tive portions of the vermis is designated as the flocculonodular bundle (Fig. 15). Fibers that are macroscopically visible enter the bundle from the parafloccular stalk, and the flocculonodular bundle in turn radiates out into the nodulus and the uvula (Fig. 15). In an effort to trace the general direction of these fibers several dissections were completed under dissecting microscope beginning with the flocculus and paraflocculus and proceeding medially. 68 Fibers from the parafloccular stalk passed into the white matter of the capula and into the flocculonodular bundle. The fibers from the flocculus entered only the flocculonodular bundle. Both the flocculus and paraflocculus appear to send fibers to both the nodulus and uvula. These preliminary observations naturally need the validation of a degeneration study. ABBREVIATIONS FOR FIGURES 10, 11, 12, 13, 14, and 15 ANS — ansiform lobule CLPL - central lobe pars lateralis CP — copula pyramidis CPL - culmen pars lateralis F — flocculus FF - floccular fissure FNB — flocculonodular bundle FP - parafloccular fissure FPn. — prenodular fissure FPo. - postpyramidal fissure FPr. - prepyramidal fissure G — parafloccular groove N - nodulus P - paraflocculus PM - paramedian lobule PMS - paramedian sulcus or groove PS - parafloccular stalk Py. - pyramidal lobule r - ridge of copula pyramidis SF - stalk folium U -umfla fl pl f2 p2 f3 floccular folia p3 parafloccular folia f4 p4 f5 p5 69 70 Figure 10. Lateral View of the flocculus and paraflocculus of the lesser bushbaby (Galago). Note the relatively small size of the paraflocculus and the in- complete differentiation of the flocculus. Figure 11. Anterior view of the flocculus and paraflocculus of the lesser bushbaby. Note the intrinsic sulci of folium pl. o .- . , . .1 .:- .3 p2 4' ‘ .2‘." '.¢ - ‘3‘. . . ~ g .. ,. / 7:9... \ ‘ ’ .v , ./ ~ - . \ . age . o . V I/‘o.. . ' I ' \ . . ”v. f 7 .' "- 'Mn‘ME \ {525:} ('1' ‘. or :19." ‘ ;. t ‘0 .t I 1. ‘ ‘1' , A ' 0". . I L, 3:5 ' ..:, 1‘ . a I J . q - I 93 J. . \ , I ‘ . f. v . i "t, | I WEAL~J’£2 . my 5; I in fa 5'4 I o f". ‘. .' \ l ”'1’ 1'. {a . 3-' \ / "e g. "'1'; \ d I. . ' .. I. I 3| ,3," e'_o:. p4 . '9‘..- sh‘ Figure 12. (A) Dorsal view of the paraflocculus of the lesser bushbaby (Galago) slightly retracted laterally. Note the lateral extent of the copula pyramidis. (B) Ventral view of the paraflocculus of the lesser bushbaby (Galago). The oval hashed line (F) represents the relative position of the flocculus. Note the shallow groove, and the theoretical line dividing dorsal and ventral paraflocculi. See text for discussion. 71 Figure 13. Caudo—ventral view of the left cere- bellar hemisphere of the lesser bushbaby (Galago). Note the distinctness of the paramedian sulcus, and the superficially visible extent of the copula pyramidis (CP). 72 Figure 14. Caudo—ventral view of the left cere- bellar hemisphere of the lesser bushbaby (Galago). The lower folia of the paramedian lobule (PM) have been dissected away and the pyramidal, uvular and nodular lobule retracted to show the extent and continuation of the copula pyramidis. 73 Figure 15. Caudo—ventral view of the right cere- bellar hemisphere of the lesser bushbaby (GhZago). Most of the copula has been removed, and the para— flocculus and uvular and nodular lobules slightly retracted to show the position of the groove be- tween the F-N bundle and copula pyramidis. 74 TURAIA GLIS AND TUPAIA CHINEMSLS The flocculus and paraflocculus of T. gZis and T. chinensis are quite similar in their gross anatomy; therefore the following discussion includes both species. The flocculus and its vermal counterpart, the nodulus, comprise the flocculonodular lobe which, due to its fiber connection with certain vestibular nuclei, compose what has been termed the vestibulocerebellum (Brodal, 1967). The flocculus is directly apposed to the lateral side of the brain stem and cerebellar hemisphere, and the paraflocculus is located in a osseous cavity in the petrous temporal bone. In the pen-tailed tree shrew (Ptilocercus lowii) Le Gros Clark (1926) described this cavity as "...a capacious parafloccular fossa." This fossa is not mentioned in his work on the skull of Tupaia (Le Gros Clark, 1925). Flocculus The flocculus of Twpaia is composed of four folia which are distinctly separated from each other by three sulci. These folia are designated f1, f2, f3, and f4, and form a structure long in its horizontal axis (Fig. 7 — p. 50, 16). Folium fl is occasionally sub- divided by a shallow sulcus, while folia f2, f3, and f4 appear to be always undivided. Folium f4 has a narrow buttress of cortex extending caudally upward and under the cOpula pyramidis lateralis (Fig. 7 - p. 50). As the stalk of the paraflocculus rests on the upper surface of folia f2, f3 and f4 it slightly flattens the dorsal (upper) surface of these lobules (Fig. 19). From an anterior view the flocculus is closely applied to the brain stem and does not "lean" laterally as in 75 76 the GaZago (Fig. 9 - p. 52). The deep groove separating the flocculus from the paraflocculus is the floccular sulcus of Le Gros Clark (1932) (Fig. 9 - p. 52, 16). This groove is also filled by the ventral edge of the osseous parafloccular fossa. Paraflocculus The paraflocculus of Tupaia is a relatively large structure, a fact implied by Le Gros Clark (1926), and is made up of six main folia (Fig. 9 - p. S2, l6, 17). These folia are designated as p1, p2, p3, p4, p5 and p6, and collectively form a distinct medial concave structure (Fig. l6, 17). In the medial angle of the paraflocculus is a very narrow folium to which the term stalk folium is applied. The five main sulci separating the six folia are very deep and the medullated core of the paraflocculus is visible in the valley of each sulcus. A dorsal view illustrates the characteristic appearance and gross configuration of the paraflocculus in Tupaia (Fig. 17). It is also noted in figure 17 that portions of the ansiform lobule over- hang pl of the paraflocculus. In turn, since the paraflocculus is quite large in Tupaia, folia pl and p2 invaginate portions of crus I of the ansiform lobule (Fig. 7 - p. 50, 9 - 52, 17). This is the characteristic appearance in both T. gZis and T. chinensis. In the adult Tupaia no dorsal and/or ventral paraflocculus can be determined. It tentatively appears that folia p1 to p4 form the upper (Dorsal para- flocculus) portion and folia p5, p6 and the stalk folium form the lower (Ventral paraflocculus) portion. 'This suggestion is based on a comparison of the paraflocculi of T. glis and T. chinenses. In 77 T. chinensis the rostral three sulci are at more distinct right angles to the caudal two than the same region in T. glis. However this relationship can be seen in the paraflocculus of the latter, it is just less accentuated. This anatomical juxtaposition leaves the observer with the impression that the rostral four folia (p1 - p4) possibly have a different origin than the caudal two folia (p5, p6, and stalk folium). Folium p1 and the stalk folium have intimate topographical relationships with the parafloccular stalk and the c0pula pyramidis lateralis. The parafloccular stalk is large in the tree shrew and continuous medially with the tear—drop folium of the copula pyramidis lateralis (Fig. 18, 19). The stalk folium passes onto the stalk of the paraflocculus then rapidly diminishes in size. A narrow band of cortex is directly continuous from folium pl into the cortex of the copula pyramidis lateralis (Fig. 19). Fibers from the stalk of the paraflocculus also enter the white matter of the copula pyramidis lateralis. Thus there is clearly an intimate mor- phological relationship between the paraflocculus and the copula pyramidis lateralis. The deep cleft between the paraflocculus and the lateral hemisphere is the parafloccular fissure of Le Gros Clark (1932) (Fig. 9 - p. 52). The dorsal edge of the osseous para- floccular fossa extends into this space. In Thpaia the c0pula does not extend laterally, consequently the ansiform lobule is the most lateral portion of the hemisphere. Copula Pyramidis In the discussion of the gross topography of the Tupaia cere- 78 bellum it was recorded that what appears to be the lowest folium of the hemisphere is actually the copula pyramidis. In Tupaia the copula is differentiated into two distinct and consistent folia, designated, in this study, as the copula pyramidis lateralis and copula pyramidis medialis (Fig. 7 - p. 50, 17). The morphological relationships of the cortex and fibers of the paraflocculus to the smaller of these two folia leave no doubt that it is part of the copula. Directly medial to .the smaller folium is a narrow folium extending toward, and into, the paramedian groove. This is the copula pyramidis medialis, and it is directly continuous with the pyramidal lobule. This junction is narrow, and a groove is formed by the manner in which the copula pyramidis medialis joins the pyramid. These two folia collectively form the bridge of cortex that joins the paraflocculus to the pyrmidal lobule, the criterion used by Elliot Smith (1903) to designate the copula pyramidis. In both T. gZis and T. chinensis the copula con- sists of medial and lateral portions, separated by a moderately deep fissure (Fig. 7 - p. 50). In a study of Tupaia minor Le Gros Clark (1924) noted that the cepula "...is slightly notched transversly at its lateral extremity...", while in PtiZocercus lowii the transverse notch was not present (Le Gros Clark, 1926). It is more significant to note that in a study of a wide variety of insectivora representing "basal" and "advanced" forms the copula pyramidis was consistently observed as being usually wide and always unfissured (Le Gros Clark, 1928, 1932). The fiber connections of the flocculus and paraflocculus of 79 Tupaia await illucidation via a degeneration technique. With care, and under a dissecting microscope, gross bundles of fibers can be observed and traced. If the copula pyramidis medialis is removed and the pyramidal, uvula and nodular lobules retracted, a delicate bundle of medullated fibers is seen coursing between the flocculus— paraflocculus and uvula-nodulus complexes. This bundle of fibers is designated the flocculonodular bundle (Fig. 19). Le Gros Clark (1924) reported no grossly observable connections between the flocculus and nodulus in Tupaia minor. Larsell (1947a; 1952; 1953) terms this the floccular peduncle in the developing human, rat, cat and monkey. The term flocculonodular bundle is preferred since the term itself is self-eXplanatory, whereas floccular peduncle is not. The post— pyramidal and prenodular fissures have no grossly detectable lateral continuations. The postpyramidal fissure essentially fades into the nondescript groove between the copula pyramidis medialis and the flocculonodular bundle (Fig. 19). A macroscopic dissection of the fiber bundles beginning in the parafloccular stalk and proceeding medially reveal contributions to the white matter of the c0pula pyramidis lateralis and medialis and contributions to the flocculo- nodular bundle. Fibers arising in the flocculus pass only to the flocculonodular bundle. This bundle in turn distributes to the uvula and nodulus of the vermis. It is interesting to notice that the flocculus, uvula, and nodulus receive primary and secondary vestibular fibers (Dow, 1936; Carpenter, 1967), therefore the possibility exists that distinct association bundles connect these respective areas and 80 the paraflocculus. The above mentioned observations on the fibers in the flocculonodular bundle, and the postulation of association bundles merit the validation or refutation of a degeneration study. ANS CPL CPM FNB FPn. FPo. . FPr. FS PM Py. SF fl f2 f3 f4 ABBREVIATIONS FOR FIGURES LB, 17, 18Land 19 ansiform lobule copula pyramidis lateralis copula pyramidis medialis flocculus flocculonodular bundle prenodular fissure postpyramidal fissure prepyramidal fissure floccular sulcus nodulus parafloccular paramedian lobule pyramidal lobule stalk folium Uvula floccular folia 81 parafloccular folia Figure 16. Lateral view of the flocculus and paraflocculus of the tree shrew (Tupaia glis). Note the relatively large size of the para- flocculus, and its medially concave appearance, and the differentiation of the flocculus. 82 Figure 17. Dorsal and slightly lateral view of the paraflocculus of the tree shrew (Tupaia glis). Note the apposition of the ansiform lobule (ANS) to p1 and p2, and the position of the copula pyramidis lateralis and medialis (CPL-CPM). 83 Figure 18. Caudo-ventral view of the right half of the cerebellum of the tree shrew (Tupaia glis). Portions of the paramedian lobe have been removed and the pyramidal, uvular and nodular lobules retracted to show the medial to lateral extent of the c0pula pyramidis. 84 Figure 19. Caudo-ventral view of the left cere- bellar hemisphere of the tree shrew (Tupaia glis). The copula pyramidis medalis has been removed and the uvular and nodular lobules retracted to show the position of the flocculonodular bundle. Also note the relationship of p1 to the copula pyramidis lateralis. The asterisk marks the position of the groove between the F-N bundle and the copula pyramidis. 85 CORRELATIVE DISCUSSION A review of the literature on the gross morphology of the flocculus and paraflocculus of prosimian primates reveals little or no agreement (table I). The observations on the flocculus of the Tarsius are particularly striking. It is doubtful that the flocculus of Tarsius evolved so rapidly. A similar review of the literature for insectivores reveals less than minimal the available information on the flocculus and paraflocculus in this phylogenetically important group of animals (table II). The flocculus of the tree shrew and bushbaby are each composed of four folia, with the former being not only more distinctly differen- tiated into individual folia, but also slightly larger in relation to the remainder of the cerebellar hemisphere. Ariens Kappers et al., (1936) observed that the direct and indirect fiber contributions of the vestibular nuclei to the flocculus undoubtedly play a role in its development. They further observe that flocculo-fugal fibers indirectly effect coordination of eye movements. It is the conclusion of these authors that the degree of eye muscle coordination is directly related to the size of the flocculus in certain mammals (Ariens Kappers et aZ., 1936). It is a well known fact that the vestibular nuclei send descending fibers to the cord as the vesti- bulospinal tract and the descending limb of the medial longitudinal fasciculus (Nyberg-Hansen, 1964). The primary ascending fibers from the medial vestibular nucleus form the ascending limb of the medial longitudinal fasciculus and project bilaterally on all nuclei 86 87 .nmmmsomap no posoapcoe 902 a suma soaaaa :Owama: :Ownma: savanna Amoesmam :ouscaE= ca ozone waaom m 1 «V a sakeoemm savages mmma camaoos .sOMEmwpxo amnmae sauna :voumasnoaane m spa: Amaaomv wrong m n m assumes mmoma spasm poaaam emexeaemmaue ooma amuse: use a waaom n m emmseaemmxue smeaeo Empocuosmmmcmx c Amonsmam ca stone maaom my a mzeosoo mxuosoo mzaeomnomz coma aunsamsnmanx eaaom m waaom same a osssnas examoonowz amma xnmao mono ea uaaom s - a =maaom mo mama emasmownaz eases asses mmoma spasm poaaau :aamEm: maaom u m wees» «someooavem mama xemao moem on zmaaom aamem 3mm: waaom a - m noses senses smma xgmao mono we woqsoooqm moaaoooamHBommzH mo mDADUooam 92¢ mDADoooqm