Search

The corpus cerebelli of mormyrid fishes is strongly developed and differentiated into four lobes: C1−C4. Although all of these lobes contain the characteristic cerebellar layers: granular, ganglionic and molecular, they show distinct architectonic differences. A previous study revealed that the ganglionic layer of C1, in addition to Purkinje elements contains conspicuous giant cells. In the present paper the results of a further analysis of C1 are reported. This analysis is based on serially sectioned brains of Gnathonemus petersii, stained according to Nissl, Bodian and Haggquist. Semi-thin sections were stained with p-phenylenediamine. Routine EM techniques were used to visualize synaptic relations. Mossy fibres and granule, Golgi, Purkinje and stellate cells are located characteristically throughout C1. It appeared that the giant cells of a previous study represent the largest elements of a population which has been termed now the eurydendroid cells. The average size of the latter is somewhat larger than that of the Purkinje cells, but both groups of cells show a considerable overlap in the size of their somata. Purkinje cells and eurydendroid cells are present throughout the ganglionic layer and both have a flattened, sagittally oriented, dendritic tree that extends into the molecular layer. Yet, the eurydendroid cells (EC) display the following characteristics which distinguish them from Purkinje cells (PC): (1) In EC the Nissl substance is dispersed diffusely throughout the soma, whereas in PC it tends to be concentrated around the nucleus, (2) The soma/nucleus ratio for EC is distinctly larger than for PC, (3) The dendritic trees of EC extend over a larger stretch of the molecular layer than those of PC, hence the term EC, (4) The dendrites of EC are more widely spaced and oriented less strictly parallel to each other than those of PC, (5) The dendrites of EC are somewhat irregular in outline and not covered with spines, in contrast to those of PC, (6) The axons of EC are oriented radially and join bundles of coarse fibres which leave the cerebellum whereas the axons of PC extend and ramify within the ganglionic layer, (7) The somata of EC, contrary to those of PC, are enveloped by a dense axonal plexus which forms numerous synaptic terminals on them. The numerical ratio of EC: PC was 1:5.5. The circuitry in C1 and the possible functional roles of its constituent neurons are discussed. It is pointed out that in this lobe the axons of PC impinge on EC and that the latter constitute its output system.

In this paper the telencephalon of Latimeria chalumnae, the only surviving crossopterygian, is described and compared to that of other lower vertebrates. It is concluded that Latimeria cannot be related to a particular group of vertebrates, but stands intermediate between the dipnoans and the actinopterygians in its forebrain structure. With respect to the shape of the subpallium, the structure of the telencephalon medium, and the arrangement of its fiber systems, the latimerian forebrain closely approaches the dipnoan condition. The pallium and membranous structures of the telencephalon of Latimeria, on the contrary, are reminiscent in gross form and histological structure of their actinopterygian homologues. However, not all the structural features of the latimerian forebrain can be related to either the actinopterygian or the dipnoan plan. The subpallium, for instance, is more primitive than that of either group mentioned; in fact, it is more simply organized than that of any other living gnathostome. The forebrain of Latimeria appears to display no special structural affinities to the amphibian forebrain. This is not too surprising, since the Coelacanths, among which Latimeria is classified, represent only a side branch of the Crossopterygii, and are not in the main line of evolution to higher forms. It is known that members of the same class of lower vertebrates may vary considerably in their forebrain structure. Hence, the Rhipidistia, totally extinct Crossopterygii which are believed to have given rise to the terrestrial vertebrates, may have possessed a forebrain quite different from that of Latimeria.

This paper introduces a research project within the framework of which the brain stems of twelve representative vertebrates will be analysed according to a new procedure termed topological analysis. The first part reviews the development of the concept that the brain stem essentially consists of a number of longitudinal zones or columns. From this review it appears that, although during the past 80 years a considerable amount of evidence favoring this concept has been accumulated, a number of important questions pertaining to number, delimitation and content of the zones in the brain stem are still awaiting a definitive answer. The second part of this paper is devoted to a detailed description and a critical evaluation of the topological procedure. This procedure, which has been especially designed to facilitate the study of the zonal pattern of the brain stem, essentially involves two steps. First the cell masses are projected back to their site of origin at the ventricular surface, and next the ventricular surface, with its sulci and the projections of the cell masses marked upon it, is flattened out, i.e., is subjected to a one-to-one topological transformation. In the third part of this paper the choice of the species which will be studied within the framework of our program is discussed. The species selected are: the lamprey Lampetra fluviatilis, the shark Squalus acanthias, the actinopterygians Acipenser fulvescens, Lepisosteus osseus and Salmo irideus, the brachiopterygian Polypterus ornatipinnis, the lungfish Lepidosiren paradoxa, the crossopterygian Latimeria chalumnae, the amphibians Ambystoma mexicanum and Rana esculenta, the turtle Testudo hermanni, and the opossum Didelphis virginiana. In the final part the objectives of the project are discussed. Most important among these are: (1) furnishing clear and succinct analyses of the brain stems of representatives of all groups of vertebrates, (2) specification of the fundamental plan of the brain stem, (3) facilitation of the establishment of homologies and simplification of the nomenclature and (4) presentation of a sound basis for experimental hodological work.

The ventricular sulcal pattern and the cytoarchitectonic organization of the brain stem of the turtle Testudo hermanni have been studied in transversely cut, Nissl stained serial sections. Four longitudinal sulci, the sulcus medianus inferior, the sulcus intermedius ventralis, the sulcus limitans and the sulcus medianus superior could be distinguished. With the aid of the usual cytoarchitectonic criteria 54 cell groups have been delimited; 12 of these are primary efferent or motor nuclei, 13 are primary afferent or sensory centers; 10 of the nuclei distinguished are considered as belonging to the reticular formation. The remaining 19 cell masses can be indicated as “relay” nuclei. In order to provide a basis for experimental hodological work the topographical position of the nuclei is illustrated in photomicrographs of representative levels and in graphical reconstructions. In order to elucidate the morphological pattern of the cell masses and their relationship to the ventricular sulci the brain stem was subjected to a topological analysis. Essentially this method requires that (1) the cell masses are projected upon the ventricular surface and that (2) this surface, with its sulci marked on it and with the outlines of the subjacent cell masses projected upon it, is flattened out. This analysis yielded the following results: The sulcus limitans extends throughout the rhombencephalon, dividing this brain part into a basal plate and an alar plate. In the basal plate the sulcus intermedius ventralis marks the boundary between an area ventralis and an area intermedioventralis containing, respectively, the somatic motor nuclei of primary and higher order, and the visceral motor centers. Yet the terms somatic motor zone and visceral motor zone are not applicable to these two areas as a whole since both contain a number of non-motor centers. Thus the intermedioventral zone harbours two general somatic sensory centers, i.e., the spinal- and princeps nuclei of V. The medial parts of the tegmentum isthmi and of the tegmentum mesencephali contain somatic motor centers of primary and higher order, and thus may be considered rostral continuations of the rhombencephalic ventral zone. The rhombencephalic alar plate contains visceral sensory, general somatic sensory and special somatic sensory centers, but these do not show a simple ventrodorsal arrangement. A prominent feature is that the special somatic sensory nuclei of termination of VIII, the cerebellar nuclei, and a number of special somatic sensory centers of higher order together constitute a continuous zone which extends from the rhombencephalic alar plate, via the isthmus into the lateral part of the tegmentum mesencephali. In the isthmus region the sulcus limitans does not mark the ventral boundary of this zone, but rather passes over its middle.

Publisher Summary This chapter focuses on the comparative anatomy of the cerebellum. It presents a survey of the comparative anatomy of the cerebellum by way of introduction to this meeting which will be devoted almost entirely to the structure and function of the mammalian cerebellum. With regard to proportionate size and shape the cerebellum is doubtless the most variable part of the central nervous system, and its histological structure is also subject to considerable differences among the various groups of vertebrates. The corpus cerebelli receives its sensory input from various sources. Spino-cerebellar and trigemino-cerebellar tracts, conveying proprioceptive and exteroceptive information toward this part of the cerebellum, have been described for all groups of vertebrates. In mammals the ‘direct’ spino-cerebellar tracts are augmented by several ‘indirect’ systems, which have a relay nucleus in the medulla oblongata (n. cuneatus externus, n. funiculi lateralis, n. reticularis paramedianus). In addition to the spino-cerebellar and trigemino-cerebellar tracts, which may be termed the ascending afferents, the corpus cerebelli receives numerous fibres which descend from more rostrally situated centres.

Publisher Summary This chapter presents a survey of the comparative anatomy of the parts of the telencephalon that are entirely or predominantly related to the sense of smell. The structures included in this so-called “rhinencephalo” display a great variety in proportionate size and degree of differentiation, but nevertheless it is possible to recognize a common structural plan that underlies the morphology of this brain part in all groups of vertebrates. The central olfactory system of the various groups of submammalian vertebrates has been discussed. Attention is drawn first to the marked variations in development of the olfactory centres. It seems probable that the early vertebrates were endowed with a well developed olfactory apparatus, and that this macrosmatic condition has been maintained in the main line of vertebrate evolution. A fargoing regression of the olfactory apparatus and its centers has occurred at least 3 times in the course of phylogeny of the submammalian vertebrates, namely in the teleosts (Mormyridae), in the lizards (Chameleon, Anolis), and in birds. In all vertebrates, the primary olfactory elements are represented by neurosensory cells situated in the nasal epithelium. These elements are provided with extremely fine, unmyelinated axonal offshoots that convey the olfactory impulses to the bulbi.

IN teleost fishes the cerebellum is well developed and consists of two principal parts: the corpus cerebelli and the valvula cerebelli. The former is massive and usually arches backward; the latter is a pouch-like structure which projects forward under the tectum in the mesencephalic ventricular cavity. It has long been known1,2 that in one group of teleosts, the Mormyrids, the cerebellum attains amazing dimensions. In these forms the valvula consists of an intermediate part and two lateral lobes, and it is the latter which has greatly hypertrophied. These lobes have grown out of the ventricle of the midbrain, to become superficial structures which cover all other parts of the brain. Fig. 1 shows that the lateral lobes consist of an arched and folded basal layer on which are numerous small and regularly arranged ridges. The intermediate part of the valvula is composed of two lobules which are situated in front of the corpus. The corpus cerebelli itself is bent forward and is of average size (Fig. 1).

The investigations of His, Herrick, Johnston and many others have led to the doctrine that the central nervous system fundamentally consists of a number of longitudinally arranged zones or columns, some of which are delimited from each other by distinct ventricular sulci. In order to facilitate the study of this basic morphological pattern in the brain stem, a procedure has been developed, making it possible to survey in one figure the neuronal structures and their relationship to the ventricular grooves in this region. Essentially this procedure involves two steps: first the cell masses and large individual cells are projected upon the ventricular surface, and next the ventricular surface is flattened out, i.e., subjected to a one-to-one continuous topological transformation. Such a topological analysis was carried out on the brain stem of the lamprey, Lampetra fluviatilis, and yielded the following results: The sulcus limitans of His extends throughout the rhombencephalon, but does not enter the midbrain. The rhombencephalic basal plate can be clearly divided into a medial somato-motor, and a lateral viscero-motor zone. The somato-motor zone consists of a rostral extension of the spinal motor column and three reticular nuclei. Rostrally this zone is continuous with the tegmentum mesencephali, which contains another reticular nucleus and the superficial and deep nuclei of III. The reticular nuclei contain groups of large cells. On either side seven of these large cells, three in the midbrain, one in the isthmus and three in the hindbrain, are significantly larger than the others, and are designated as Muller cells. The viscero-motor zone is confined to the hindbrain. It consists of a continuous chain of discrete cell masses: the motor nuclei of V, VII, IX and X. This zone contains in addition the large Mauthner neuron, which is situated between the nuclei of V and VII. In the rhombencephalic alar plate, separate viscero-sensory and somatosensory zones could not be recognized. The delimitable nuclei in this region all belong to the somato-sensory octavo-lateralis complex. The latter complex extends rostrally into the very small cerebellum which, apart from a zone of periventricular gray and a nucleus cerebelli, harbours the motor nucleus of IV. The sensory centers of the midbrain, i.e., the torus emicircularis and the tectum are caudally continuous with the gray substance of the alar plate of the hindbrain. Large, presumably primary sensory, elements occur in the caudal and ventral part of the rhombencephalic alar plate. Such “dorsal cells” are also present in the spinal cord, but a midbrain equivalent, i.e., nucleus mesencephalicus of V, is entirely lacking.

The diencephalon of Polypterus can be divided into an epithalamus, thalamus, and hypothalamus. The habenulae, the nervous parts of the epithalamus, are comparable to their homologues in other lower vertebrates with respect to sulcal boundaries, cellular structure, and fiber connections. The thalamus of Polypterus is not divisible into a pars dorsalis and pars ventralis by the sulcus medius; rather this sulcus is in the middle of a uniform, laminated cytoarchitectonic field. In this respect Polypterus differs from other species in whom the sulcus medius divides the thalamus into dorsal and ventral parts. There are six migrated nuclei in the thalamus of Polypterus. There is only one circumscribed projection into the thalamus, i.e., the optic tract, but there are numerous diffuse fibers terminating in this region of the brain. The hypothalamus, except for a partially migrated nucleus, has retained the periventricular arrangement of cells. It has large fiber connections with the forebrain and brainstem. The literature on the diencephalon of lower forms has been reviewed with special emphasis on the question of how homologies are established in this brainpart. It appeared that three different criteria, either singly or in combination, have been employed as a clue to identification of structures in the diencephalon. These are, (1) ventricular grooves, (2) nuclear boundaries, and (3) fiber connections. In order to test the practical validity of these criteria the diencephalon of Polypterus was compared to that of five related species, i.e., the actinopterygians Acipenser and Polyodon, the dipnoans Protopterus and Neoceratodus, and the crossopterygian Latimeria. In addition three amphibians, Necturus, Ambystoma and Rana, were involved in our comparative considerations. It was concluded that, within the confines of the diencephalon of the species mentioned, cytoarchitectural differences form the most valid criterion for establishing homologies. The drawback and restrictions connected with the use of ventricular sulci and fiber connections, as a clue to identification have been evaluated and discussed.