In the earliest edition of The Origin of Species Darwin devoted Chapter XIII to 'Mutual Affinities of Organic Beings: Morphology: Embryology: Rudimentary Organs' and gave clear accounts of the importance of these disciplines to his studies in natural history [1]. Of morphology he wrote "This is the most interesting department of natural history, and may be said to be its very soul" and of embryology "This is one of the most important subjects in the whole round of natural history". Such authoritative statements encourage this presentation of studies of the morphology and embryology of the cerebrospinal fluid (CSF) system as a contribution towards the understanding of its evolution; they also provoke a preliminary consideration of the relationship between morphology, embryology and evolution in Darwinism as a whole.
The detailed descriptions of the remarkable journey around South America while Darwin was naturalist aboard HMS Beagle(1832-1837) and the extensive records of his observations upon the flora, fauna, and geology [2] leave no doubt that Darwin himself was a comprehensive morphologist as well as a great natural scientist. His Introduction to the first and subsequent editions of The Origin of Species explains how his early thoughts on the subject arose from his observations on the distribution of 'organic beings' made while he was on HMS Beagle and that he did sketch out earlier versions of his theory in 1842 and 1844. These essays were fortunately preserved and published in their original form by his eldest son, Francis Darwin; they confirm that Darwin's basic approach to the presentation of his work was established early[3]. This approach consisted of: firstly, the study of variations under domestication and the development of these by inheritance and breeding until they virtually become distinct species; secondly, comparable variations which occur in the wild state are selected out by natural means if they are advantageous in the struggle for survival, and thus a new and successful species originates; thirdly, the evidence for this process having occurred is found in geology and palaeontology, and the geographical distribution of species. Darwin then paid great attention to the various affinities among species which provide the morphological basis for their classification, and the structural evidence that they came from ancestors within the great animal phyla and classes. He considered the embryological development of a species reflects its origin and place among the phyla - a concept expanded by later authorities [4,5].
In both essays there is a third section in the study of variations which is concerned with those which occur in instincts and mental behaviour, both under domestication and in the wild, and in considering the difficulties presented to the theory of natural selection by these aspects of variation Darwin writes: "one is forced to admit that mental phenomena (no doubt through their intimate connection with the brain) can be inherited like infinitely numerous and fine differences of corporeal structure" [3]. Thus the place of the brain as a determinant of behavioural variations which become important in the struggle for survival was considered with his early thoughts on the origin of species.
The same Introduction to the 'Origin' referred to above accounts for the years from 1837 to 1859 as having been spent in consolidating the observations confirming his theory of Natural Selection as the means whereby species originated, and, as is well-known, the publication of the 'Origin' in 1859 was precipitated by receiving the paper from Wallace on the same subject and the presentation of their joint paper to the Linnaean Society in 1858 [6]. Darwin regarded The Origin of Species as very much an incomplete account of his work - an Abstract as he called it - and these facts may explain the lack of references, particularly in the first two editions; the text clearly reflects Darwin's wide knowledge and reading on the topics detailed in each chapter, but it was only in the 'Historical Sketch' added to the American edition and the 3rd English edition that Darwin formally acknowledged prior observations and theories indicating that species of animals may have arisen other than by specific acts of creation [7,8]. In this 'Historical Sketch' he quotes the works of such authorities as Lamarck, St.Hilaire, Owen and others, adding the somewhat wry footnote towards the end that of the thirty-four authors quoted who 'believe in the modification of species, or at least disbelieve in separate acts of creation, twenty-seven have written on special branches of natural history or geology'.
Morphologists, and in particular, Professor Richard Owen who had continued the tradition of John Hunter at the Royal College of Surgeons to become one of the greatest authorities on both vertebrate and invertebrate structure, were prominent in Darwin's historical sketch, and his indebtedness to them is clear. His contention, particularly in relation to Owen, was that they observed and compared morphology, looking for homologues, but were unclear as to how or why such relationships occurred. Owen had not only been one of the authors responsible for the official presentation of the Zoology of the Voyage of the "Beagle" [9] but had also published his own detailed lectures on the comparative anatomy and physiology of both vertebrates and invertebrates before 1859 [10]. In the same year as the first edition of the 'Origin' Owen wrote in an address on 'Homo in relation to mammalian classification by brain structure': "The supreme work of Creation has been accomplished that you might possess a body - the sole erect - of all animal bodies the most free - and for what? for the service of the soul" but adds to this work an appendix "on the gorilla, and on the extinction and transmutation of species" [11]; it was the juxtaposition of such contrary statements that led Darwin to express some difficulty in understanding the position of this eminent morphologist on the origin of species.
The contents of the early editions of the 'Origin' are arranged under the same headings as the 1844 essay and summarised in the full title 'The Origin of Species by means of Natural Selection or the Preservation of Favoured Races in the Struggle for Life' and it is of interest to notice that in the considerably modified 6th edition of 1872 the chapter on instinct is placed among those dealing with 'Objections to the Theory of Natural Selection' and is concerned more with instinctive group behaviour among insects then extrapolated to tribal sociology rather than with the brain-behaviour relationship [12]. Furthermore, even in this later edition, the index does not actually include the word evolution, and only on its penultimate page does Darwin reflect upon the origin of man and the past and future of species as a whole. The 'Origin' thus remains the Abstract it was introduced as, and presumably The Descent of Man and Selection in relation to Sex first published in 1871 should be considered Darwin's major work[13]. Certainly it is this book which firmly places man as the ultimate outcome of the evolution of species by a process wherein natural selection of advantageous variations consists primarily of sexual selection of physical, intellectual and moral faculties; it is noteworthy that the second edition of this book in 1874 has appended Huxley's 'Note on the Resemblances and Differences in the Structure and Development of the Brain in Man and Apes'[14] thus relating this particular aspect of Darwinism to the morphology and embryology of the brain and thereby setting the stage for many subsequent studies.
Domenico Cotugno [15] gave the first clear account of the water which filled the space between the dura and spinal cord, and it seems likely that previous descriptions of vapours and spirits around the brain were based on less careful dissections which permitted the Cerebrospinal Fluid(CSF) to drain away and be replaced by air or the gases of putrefaction. The recognition of the CSF as a system must be attributed to Francois Magendie who contributed a series of studies extending from 1825 to 1842 [16,17]. He noted the continuity of the intracranial with the spinal fluid, and of the fluid around the brain with the fluid within the ventricles through a midline foramen in the roof of the fourth ventricle, which is commonly called by his name. He considered that the external fluid was formed on the surface of the brain and the intraventricular fluid was formed by the vascular frond-like structures within the ventricles, known as the choroid plexuses. He postulated that there was a free exchange between the internal and external fluid by an ebb and flow through his foramen, and that the function of the CSF was to protect the brain and spinal cord and to float them within their bony container.
Magendie's observations appear to be based upon human subjects, as do those of Luschka describing the lateral foramina of the fourth ventricle, and those of Key and Retzius describing the relationship of the subarachnoid space to the arachnoid granulations [18,19]. However, these observations do not appear in the monographs on vertebrate morphology by Owen and Huxley in the latter half of the nineteenth century [20,21]; their accounts lack detail of the structure of the fourth ventricle roof and the meninges, and do not mention the CSF as a system.
In this same period following the impact of Darwinism, and doubtless due to the continued emphasis on the importance of the role of the central nervous system in the evolution of man, works devoted to its morphology and evolution appeared but these do not include separate or detailed consideration of a CSF system [22,23,] and the few communications on related structures are limited to consideration of one component, and few species of animals [24,25]. Nevertheless, the reports that teleost fish have no arachnoid [27] and that the midline foramen (or metapore) in the roof of the fourth ventricle can be found in man and several Old World monkeys, but not in the cat [28] might have been expected to arouse the interest of the comparative morphologist had the CSF system been recognised as such, and its evolution considered separately.
The study of any system consisting of fluid in and around a structure such as the neuraxis and contained by fine membranes, presents considerable technical difficulty when the classical method of serial section histology is used, and this may be one reason for its continued neglect. A second reason almost certainly is the concept of a CSF circulation which developed during the first two decades of the twentieth century from the observations of Weed [29,30] using potassium ferricyanide injections, and those of Dandy and Blackfan[31] upon the production of hydrocephalus. When Cushing gave the Cameron Lectures in 1925 [32] he found enough evidence from the work of his colleagues and that of his own on surgical cases to put forward the concept of a 'cerebrospinal circulatory sector' or 'third circulation' consisting of: the secretion of fluid from the choroid plexuses, its passage through the ventricular system and out of the foramina in the roof of the fourth ventricle, through the basal cisterns and the subarachnoid space to its site of absorption at the arachnoid granulations of the superior sagittal sinus. This concept has become classical and has dominated the understanding of the CSF ever since, although it has never fully satisfied some clinical observations in humans, and many observations derived from experimental work. Furthermore, this concept of the CSF system does not account for its behaviour in animals which lack many of the structural features found in mammals.
In bibliographic terms de Beer can be considered as providing direct continuity with Darwinism in the disciplines of morphology and embryology, the first edition of his book on vertebrate zoology having been supplied with an introduction by T Huxley's son, the zoologist JS Huxley[33]. It contains descriptions of some aspects of the CSF system in animals from amphioxus to the rabbit (in which he describes the arachnoid mater and a foramen of Magendie) but not in full detail for all of the animals studied, nor in the context of a system as such, and the account of the CSF-related structures remained the same in the second edition some forty years later[34] but his further work on embryology elucidated its position in relation to phylogeny and the so-termed 'Law of Re-capitulation'[5,35]. Among other twentieth century authorities on zoology Romer in his first and subsequent editions[36] describes the meninges of fishes as a single layer of compact tissue, and states that all tetrapods have an outer dura mater, and an inner layer derived from the neural crest, the latter being divided into arachnoid and pia in mammals; he describes the choroid plexuses of the third and fourth ventricles, but makes no mention of ventricular foramina or CSF flow. Weichert, in his first edition[37] describes the cyclostomes and fishes as having a single-layered meninx (meninx primitiva) with peri-meningeal spaces, and urodele amphibians as having two more-or-less distinct layers - the inner pia-arachnoid and outer dura mater. He describes the three layers in mammmals, and the CSF as passing through three openings in the roof of the fourth ventricle; he does not account for the CSF system in the other animal phyla, and his descriptions remain unchanged in four subsequent editions. Eaton's book on the comparative anatomy of vertbrates[38] describes fishes as having a single meninx, and amhibians, birds, and reptiles as having an outer dura mater, and an inner pia mater close to the brain; the third meninx, the arachnoid, appears in mammals, enclosing the CSF in the subarachnoid space. Kent, also writing on comparative anatomy, gives no comparisons of CSF structures, but describes a 'sluggish circulation' out of the fourth ventricles to sites of absorption along the roots of the spinal nerves to the lymphatic system, as well as into the venous system via the arachnoid granulations. In his earliest edition[39] he states that the CSF system functions as a protection to the brain, as well as a means of exchanging metabolites and neurotransmitters, while in the third edition he goes so far as to say 'the fluid assists in protecting the central nervous system from concussion' [40]. Montagna describes the Cushing classical CSF pathway, and notes the limited meningeal differentiation in fishes[41], and the authoritative account of the vertebrates by Young[42] emphasizes the study of behaviour as well as structure in the evolution of species, and culminates in his classical account of the evolution of the nervous system [43], but in his detailed presentation, including the phylogeny of the endo- and peri-lymph components of the ear, there is almost no attention to the CSF.
In the last three decades zoologists have reflected interest in the CSF system as a means of maintaining the levels of extracellular metabolites, and have related this to species differences in the ventricular system and other ependymal-derived structures[44]. Kappers extended his earlier interest in the comparative morphology of CSF-related structures to that of the ependymal derivatives also, and included these in a comprehensive text on the comparative anatomy of the brain [45]; a similar emphasis is contained more succinctly in the small volume by Sarnat and Netsky, the second edition of which contains the most interesting observations upon the activities of the wood-pecker and the particular CSF structure of birds[46,47].
Following Cushing's presentation of the CSF system as the 'Third Circulation' it became a topic for monographs presenting its physiology and the laboratory analysis of the fluid as a means of diagnosing diseases of the central nervous system[48,49,50,51,52,53] but the observations and experiments contained in these accounts are derived from humans or other mammals - the comparative morhology of the system is poorly represented. The slim volume by Millen and Woollam[54] remains a classical and concise account of the anatomy of the cerebrospinal fluid in the human, but it contains no account of its comparative morphology nor its multiple functions.
Of these functions it is undoubtedly the role of the CSF in the homeostasis of the extra-cellular environment of the central nervous system which has predominated in the basic scientific studies of the last twenty-five years. Among these the notable contributions of Davson to the study of the CSF as a system, and his collation of those of other experts in the field have been presented, with co-authors, in a number of editions showing a varying emphasis around the main theme of homeostasis of the central nervous system[55]. The relationship of the CSF to the extracellular fluid and the blood brain barriers has remained central in his accounts, but intracranial pressure and pulasatile flow of the CSF is quite fully presented in the last edition[56] and the ontogenetic aspects of the CSF system, including some of the original studies to be presented here, are given an extensive chapter. The subsequent chapter, devoted to the comparative physiology of the CSF, also includes many observations derived from, and related to some of the crucial original studies upon the establishment of communication between the ventricular system and the subarchnoid space in amphibians to be detailed here. The context of these, by inference, is the phylogeny of the CSF flow pattern, but this is not analysed in relation to any protective function of the CSF.
There appears to be a place for the presentation of a full study of the comparative morphology of the CSF system in order to provide not only the structural context that will ensure dynamic studies are species-related, but also as a basis for understanding more of the functions of this system in the support of the evolving brain.
In a study of comparative morphology it is desirable to examine not only representative species from each class of vertebrates, but also representatives of different orders and even different genera in order to establish the overall pattern of similarities and differences. Futhermore, while the full study of the morphology of the CSF system in one specimen of any particular species is probably of significance for that species, it does not exclude the possibility of individual variations and pathological abnormalities; observations based upon five or ten specimens could well have an increased value. For example, the experience of examining large numbers of specimens within the class Amphibia revealed important differences between those of the orders Anura and Urodela, but the morphological features in general were the same within each order, and showed only minute species variations.
Most studies of the structure and function of the CSF system hitherto have been made in mammals and, generally, the differences between species is disregarded; the studies which form the basis of this presentation are concerned with the vertebrate kingdom as a whole and the relatively major structural differences which have been found are probably of significance in the evolution of the CSF system.
The structures of the system in the human adult have been well described in modern classical texts already referred to[53,54,55,56].
The morphological studies on the amphibians were of such interest and apparent significance that they were extended to the tadpole and through metamorphosis. Some original studies of the CSF system in Man included in this work are also of its embryological development.
Fixation All of the specimens were obtained live and the majority were anaesthetised by immersion in a solution containing MS222(Tricaine methane sulphonate, Sandoz) but the birds and mammals were anaesthetised with a calculated dosage of Nembutal. Fixation of the majority of specimens was obtained by intracardiac perfusion with initially Ringer's solution, isotonic for the particular species, followed by dilute Karnovsky solution [57]. The amphioxus species and the series of frog tadpoles were fixed by immersion in Karnovsky solution. In some species, when initial examination showed fixation to be unsatisfactory, the specimens were decapitated, part of the skull vault excised and the whole immersed in Bouin's solution for 24 hours. In a small number of lampreys and amphibians the ventricular system was injected with either Dextran Blue or the pigment 'Aquadag' in gelatin in order to demonstate the CSF spaces. Injections of the ventricular system with Batson's solution were performed in fresh-frozen dogfish specimens and anaesthetised amphibians; after maceration with concentrated, boiling potassium hydroxide, ventricular casts were obtained.
Decalcification Specimens were immersed in solutions of RDO for 24-48 hours for decalcification.
Clearing The lamprey and majority of the amphibians were cleared in cedar wood oil, but all of the remaining specimens were cleared in terpineol.
Tissue processing On the first day 50%,60%,and 70% alcohols were used during the day, and 80% alcohol overnight. On the second day 90% and 95% alcohol followed by three changes of 100% alcohol, alcohol and cedar wood oil in a 1:1 mixture for one hour, and pure cedar wood oil overnight. On the third day cedar wood oil and toluene, 1:1 for one hour followed by pure toluene for one hour.
Embedding The majority of specimens were embedded in wax using three changes and then left overnight in the fourth change of warm wax. They were finally evacuated in fresh warm wax to remove all air bubbles and then embedded in paraffin wax or paraplast.
Section cutting Serial sections were cut in a transverse plane at a thickness of 6um for smaller specimens but in specimens which were large and easily fragmented the sections were cut at 8um. A few specimens were sectioned longitudinally.
Staining The specimens were mounted on slides serially, and stained in rotation using the standard Haematoxylin and Eosin stains for general orientation and identification of cell types, Van Gieson's stain to demonstrate collagen in the membranes surrounding the brain and in the connective tissue within the neuraxis; the third slide in the rotation was stained with Masson's trichrome which shows collagen, reticulin, amyloid and mucin as green, nuclei as black and muscle, red blood cells and most cytoplasmic granules as red, or Mallory's trichrome with which the collagen, reticulin and basophilic substances are a deep blue and other structures, by contrast, light red or orange, and the nuclei stain purple. In a few series the trichrome stain was Martius scarlet-blue which stains connective tissue blue, fibrin red, and nuclei purple to black.
Mounting The slides were mounted using Histamount.
The amphioxus proved initially difficult to fix and section without shattering, and the most satisfactory specimens were obtained after immersion in sea water, followed by immersion in Karnowsky solution over 24 hours. Full serial sections satisfactory for study were obtained from four specimens.
The lamprey specimens were satisfactorily fixed by intracardiac perfusion with Karnovsky solution and serial sections were taken of the body and tail sections in some specimens. Six very satisfactory series were obtained, including one sectioned longitudinally.
The one dogfish obtained in the live state was well fixed with intracardiac perfusion of Karnovsky solution followed by immersion of the rather large head after the cranial vault was removed. The sections were cut at 8um.
Seven goldfish specimens were prepared suitable for study, including one which was longitudinally sectioned.
Two of the lungfish were prepared and one of these, sectioned at 10um, produced a good series.
Over 100 specimens of Rana temporaria were prepared, including a special series of 10 adults sectioned for detailed study of the roof of the hindbrain, and two series of tadpoles at various stages of development up to matamorphosis. Specimens of the following Anurans were also studied: Bufo bufo, Xenopus, Rana pipiens, Hyla arborea, Rana catasbiana, and Rana ridibunda.
Five adult salamanders were prepared and one of these was longitudinally sectioned. The tail was studied in some specimens. In addition the following Urodeles were studied: Triturus vulgaris and Necturus maculosa.
Two turtles and nine lizards were used in the study. In preparing the turtles, post-fixation of the excised head in Bouin's solution was performed, and after decalcification the specimens were softened in phenol for two days.
A similar modification of technique was used for the chick and pigeon; four birds were fully studied by serial sections and an additional pigeon was prepared and the various membranous layers at the cerebello-medullary junction were dissected and examined.
The mouse also had post-fixation in Bouin's solution after decapitation, removal of the cranial vault, and preparation of the brain by an initial longitudinal section between the cerebral hemispheres, followed by transverse sections from the midbrain caudad.
Any serial section light microscopy study of the brain and its surrounding membranes in situ within the skull vault is a demanding and time-consuming technique which imposes a limitation on the numbers which can conveniently be studied. The differential shrinkage between brain, surrounding meningeal membranes, and skull, which occurs during fixation, is likely to be reduced and the essential fixation of the neuraxis itself to be more successful following intracardiac perfusion of a glutaraldehyde-containing fixative such as Karnovsky solution; for this reason animals were not subjected to the serial section study unless they were obtained live in the laboratory.
The size of the individual specimens was also a limiting factor, and part of the technique was to trim the skull vault and base after fixation in order to obtain a more manageable block. Dehydration and embedding of these preparations containing fluid-filled spaces separating the various structures was difficult to achieve without disrupting delicate membranes, and the cutting of such preparations was only successful at thicknesses of 6um or, in some cases, 8um or 10um. These sections proved appropriate for the histological assessment of the membranes.
Using these full serial section techniques preparations were obtained in which structures could be examined in their full extent and the morphological continuity or otherwise of a membrane could be distinguished from artefacts by tracing the detailed appearance over a number of sections. In general the perfusion fixation technique caused distension of intracerebral vessels, but did not disrupt the capillary walls. Differential shrinkage was most marked between the brain and the surrounding dura, producing a subdural space which was largely artefactual in many species. The arachnoid, when present, tended to follow the neuraxis to which it is attached at various points.
The preparations, although primarily designed to maintain relationships of membranes and cavities, were of sufficient quality to identify also single-celled structures at the higher magnifications of light microscopy, and to permit examination of the fine structure of the membranes and epithelia which was regarded as essential preliminary background to ultrastructural studies undertaken in our own laboratory and elsewhere, in order to elucidate membrane morphology in greater detail.
The serial sections were fully studied with specific attention to the following structures: ventricular system, roof of the hindbrain, meninges, ependyma, choroid plexuses, and the relationship between the CSF cavities and the cerebrovascular system. A detailed description under these headings was assembled for each species studied, and alpha-numeric references made to the relevant slides and sections. The laboratory record thus obtained provided the working basis of the descriptive presentation in Section 2 of this work where the account has been limited to the salient features under each heading.
Illustration of such a comprehensive survey by use of the serial section slides is of necessity limited and arbitrary. A group of figures for each species has been assembled to illustrate the structures and relationships at different levels of the neuraxis. A further series of coloured diagrams presents the whole study for easy comparison. In addition the results are tabulated (Table III).
| Click here to return to Evolution of the CSF System - Synopsis |