The Origin and Development of the Nervous System

SCIENCE SERIES, established by the Trustees of the University, owes its origin to a belief that there should be a medium of publica- tion occupying a position between the technical journals with their short articles and the elaborate treatises which attempt to cover several or all aspects of a wide field. The volumes of the series will differ from the discussions generally appearing in technical journals in that they will present the complete results of an experiment or series of investigations which previously have appeared only in scattered articles, if published at all. On the other hand, they will differ from detailed treatises by confining themselves to specific prob- lems of current interest, and in presenting the subject in as summary a manner and with as little technical detail as is consistent with sound method. They will be WTitten not only for the specialist but for the educated layman.

Concerning the embryonic development and the evolution of nervous structure a large body of fact and hypothesis is at hand, but the problem of the origin of the nervous system as an organ of excitation, transmission, and integration has received much less attention. Irri- tability, that is, excitability, is commonly regarded as a fundamental property of living protoplasm, and it is often asserted that the nervous system does not repre- sent a new function, added at a certain stage of develop- ment, or of evolution, to the original functional complex of the organism, but rather a development of the primi- tive mechanisms of excitation and transmission. Not infrequently, however, particularly within recent years, it has been maintained that the primary factors in physiological correlation and integration consist of chemical substances and their transportation by one means or another. These two conceptions are difficult to reconcile, for the one implies that excitation and transmission are primary factors in physiological corre- lation, the other that they are of secondary origin. As a matter of fact, we know that excitation and transmis- sion occur in protoplasm in the absence of anything which we can identify as nervous structure; moreover, it is difficult to conceive a living organism incapable of at least some degree of excitation and transmission.

Considered from a physiological viewpoint, the origin of the nervous system must be sought in conditions present before the appearance of definitely recognizable morphological structure of nervous character. This book is first of all a consideration in the light of recent experimental investigation of certain of the physiological conditions which antedate the appearance of the nervous system, and with which its appearance and development appear to be closely associated. It is, in fact, an attempt to establish the existence of physio- logical continuity between the simple quantitative gradient in physiological condition and the nervous system. As I shall endeavor to show on the basis of experimental data, the physiological gradient arises in the final analysis as an excitation-transmission relation in response to the differential action of an external factor and at the same time represents the primary integrating factor of the sort characteristic of the organism. The nervous system represents the organ of integration par excellence, and some sort of continuity must exist between the primary factors concerned in determining those relations between regions or parts which make the organism an orderly whole and the nervous system, which continues the development of this integration to its highest point. Since the book is concerned first with this question of continuity and its nature and since no general consideration of the physiological gradients has appeared since 1915, it has been found necessary to devote the earlier chapters to a discussion of pattern in the organism and its relation to the gradients. These chapters constitute the basis for the consideration of the nervous system in the stricter sense, which follows. Believing as I do that the conception of the physiological gradient enables us to throw some light, not only on the problem of development in general, but particularly on

the problem of the nervous system in its physiological aspects, it has seemed a task worth undertaking, even though our knowledge is still fragmentary in many respects, to attempt to point out the relations between the gradients and the nervous system. As regards many points, of course, only hypotheses or suggestions are possible, but if these shall stimulate further investi- gation and so aid, even indirectly, in advancing our knowledge, their chief purpose will have been attained, whether they are confirmed or shown to be incorrect.

Since many different lines of investigation concerned in the discussion are of necessity mentioned only very briefly, it has seemed advisable to give references, so far as it could be done without unduly breaking the continuity of the text. The list of titles following the text makes no pretensions to completeness, but includes merely the publications to which reference is made. I am deeply indebted to certain of my colleagues and co-workers in various ways: first of all to Dr. C. J. Herrick, who has played no small part in the develop- ment of the book. While it is by no means my intention to shift to his shoulders any of the responsibility for conclusions drawn or suggestions advanced, I feel it a great pleasure, as well as a duty to acknowledge my obligations to him for many fruitful suggestions made in the course of personal discussions during the years of our association in Chicago, for information on many questions of fact, for his kindness in reading and criti- cizing the entire manuscript, and finally for permission to reproduce certain figures. I am also under great obligation to Dr. L. H. Hyman of the Department of Zoology for reading and criticism of the manuscript

and for permission to use unpublished data, and to Dr. A. W. Bellamy of the Department of Zoology for permission to use unpublished data and to reproduce certain figures. For permission to reproduce published figures, acknowledgments are due further to the following pub- lishers and authors: D. Appleton and Co.; Henry Holt and Co.; P. Blakiston's Son and Co.; The University of Chicago Press; Professor S. Ramon Cajal; Dr. F. R. Lillie; Dr. R. G. Harrison; Dr. W. A. Locy; Dr. J. B. Johnston; Dr. G. H.Parker; and Dr. Charles Hill. The sources of all figures reproduced are given in the legends. To the artist of the Department of Zoology, Mr. Kenji Toda, my thanks are due for his beautiful redrawing of

Pattern and Material. Organismic Pattern. The Origin of Organismic Pattern. Physiological Correlation as the Mechanism of Integration. Organismic Pattern and the Chromosomes. Conclusion. The Physiological Gradient. The Evidence from Struc- ture and Development. The Evidence from Suscepti- bility. The Evidence from Rate of Penetration. The Evidence from Certain Oxidation-Reduction Reactions. The Evidence from Bioelectric Phenomena. The Evi- dence from Oxygen Consumption and Carbon Dioxide Production. Evidence for the Existence of Physiological Gradients in Various Organs.

The Origin of Polarity. The Origin of Symmetry. The Problem of Cell Pattern. The Relations of Surface- Interior to Axiate Pattern. Conclusion. The Excitation Gradient. Theories of Excitation and Transmission. Primitive Excitation Processes. The Question of Reversibility. General Discussion. The Physiological Gradient as a Basis for Differentiation. Physiological Dominance and Subordination hi Relation to the Axial Gradients. Physiological Isolation. Conclusion.

The Nervous System as a Product of Physiological Factors. The Localization of Nervous Structure. The Early Differentiation of Nervous Structure. Physiological Independence and Dependence in Develop- ment in Relation to the Gradients. The Physiological Independence of the Apical Region and Head in Experi- mental Reproduction. The Relative Independence of the Nervous System in Development. The Significance of the Facts. Centralization in Relation to Axiation. Progressive Centralization and Cephalization in Various Groups. The "Stepladder" Type of Nervous System in the Invertebrates. The Position of Peripheral Receptors. Conclusion.

The Neuron. The General Morphological Character- istics of Neuron Pattern. Neuron Pattern as a Case of Axiate Pattern. The Possible Factors Concerned in the Axiation of the Neuron. Kappers' Theory of the Origin of Neuron Pattern. Electrical Polarization as a Localizing and Orienting Factor. Axon Development. Dendrite Development. The Neurons of the Spinal Ganglia. Kappers' "Neu- robiotaxis" of the Cell Body. Certain "Monopolar" Neurons. The Cells of the Nervous Reticulum. Con- clusion.

The Development of Excitation and Conduction. Trans- mission in Certain Non-nervous or Neuroid Paths. Transmission in the Row of Swimming Plates of the Ctenophore. Certain Axiate Organs as Physiological Gradients. The " All-or-None " Principle in Nervous Conduction. Conduction in the Neuron and the Prob- lem of the Synapse. The Physiological Characteristics of Reflex Arc and Excitation Gradient. Historical and Critical. The Functional Direction in the Nerve Net. Functional Direction in Certain Bilateral Invertebrates. Func- tional Direction in the Higher Animals. The Problem of the Origin and Development of the Cerebral Cortex.

FIGS. 1-3. Ovarian Eggs of Jellyfish, Phialidium gre- garium, Indicating Relation between Polar- ity and Position 55 FIG. 5. Semi-diagrammatic View of Ovarian Egg of Frog, Indicating Relation between Polarity and Circulation 5 6 FIGS. 8-12. Reorganization in Pieces of Planar la dorotoce- phala from Different Body Levels in Relation to Presence or Absence of Head . . . . 101 FIGS. 13-15. Axial Gradients in Development of the Cen- tral Nervous System 109

FIGS. 1 6, 17. Transverse Sections of Chick Embryo at Dif- ferent Levels, to Indicate Developmental Gradients m FIGS. 30-36. Partial Structures from Short Pieces oiPlanaria 125 FIG. 37. Regions of High Susceptibility in Early Larval FIG. 38. Unsegmented Egg of Frog, Viewed from Apical Pole, Showing Differential Suscepti- bility. FIG. 39. Stages of Disintegration of Frog Embryo Dur- ing Late Cleavage 137 FIGS. 40-44. Embryonic Segmentation of the Central Nerv- ous System of Vertebrates 139

FIG. 50. Monopolar Neuron from Mesencephalic Nu- cleus of Fifth Cranial Nerve 160 FIG. 53. Neuron from Ventral Cord of Earthworm . 162 FIG. 54. Development of the Pyramidal Neuron . . 163 FIG. . 56. Two Views, Taken Fifty Minutes Apart, of Same Nerve Fiber, Growing from Group of Embryonic Spinal Cord Cells into the Lymph of Tissue Culture 168 FIG. 57. Diagram Illustrating Kappers' Conception of the Action of the Bioelectric Current on the Neuroblast 176

FIGS. 62-65. Developmental Stages of Neural Tube and Spinal Ganglia in Salamander, Amblystoma punctatum 197 Life manifests itself in the form of more or less clearly denned and delimited entities which we call individuals. Biologists are in general agreed that some sort of order, unity, harmony or wholeness as it is variously called, in short, some sort of integrating or ordering factor or principle, effective in both space and time, underlies organic individuality. An adequate conception of physiological integration must afford a basis not only for the spatial localization of parts and for the sequence of events in time in the development of the individual, but also for the functional relations of both nervous and chemical character between parts. And particularly it must account for the origin and development of the nervous system as the organ of integration par excellence. The present book is an attempt to show at least that the nervous system does not represent a new integration imposed in some way upon primitive protoplasm, but rather a product, a result of the primary integrating factors which make the organism an orderly whole.

Like any other orderly entity the organic individual presents two great problems, the problem of pattern and the problem of material, and it is with the first of these, the problem of pattern, that we are here chiefly concerned. The crystal is not a purely fortuitous aggre - gation of molecules, the house is not simply a pile of lumber,.~bricks or stone, mortar, etc., nor is the organism simply a kaleidoscopic mixture of substances. In all these types of individuality the relations of the con- stituent materials are ordered and controlled in accord- ance with some sort of pattern. Pattern then, in the broad sense, is the particular kind of order, unity, or wholeness which constitutes the entity or individual, whether it be crystal, house, or organism. Moreover, in the organism as elsewhere, the distinction between pattern and material is clearly indicated by the fact that very similar patterns may exist in very different materials. For example, certain hydroids resemble very closely certain plants as regards the order of budding and the growth relations of branches. As regards general axial relations certain bilaterally sym- metrical plants, e.g., Marchantia, are very similar to certain bilaterally symmetrical animals, such as Planaria, and very similar radiate arrangements of parts are found in many different plants and animals. Again, the so-called law of anteroposterior development calls attention to a feature of developmental pattern common to many widely different organisms, and finally the receptor-conductor-effector pattern in some form, ran- ging from a momentary relation in which the three com- ponents are completely interchangeable, to the reflex arc associated with a permanent nervous structure, is appar- ently a physiological pattern characteristic of living pro- toplasm in general in relation to an external environment. Pattern develops in or is imposed upon material of some sort, and the problem of the constitution of this

material is the other great problem involved in indi- viduality. In the case of the organism this problem includes the physicochemical constitution of protoplasm in the broadest sense. This problem of material resolves itself again at each step into new problems of pattern and material. For example, the cell is the material of which multicellular organisms are composed, but the cell involves a problem of pattern and of material. The cell consists of protoplasm, and the two problems of pattern and material exist for protoplasm as well as for the multicellular organism and the cell, and not only for protoplasm, but for each colloid particle, each proteid, each molecule, and, if we conceive the term "material' broadly, each atom. This is merely a statement of the fact that entities or individuals of certain kinds or degree of complexity, each with its own pattern and material, may be integrated to form more complex individualities which again may serve as material for a new integration and so on. The problem with which we are here primarily concerned is the problem of the nature and origin of the pattern which constitutes the organism as a whole, whether it consists of one cell or many, in short, the problem of organismic1 pattern. While pattern constitutes the basis for an ordering, an integration of the material, it must neces- sarily be of a sort that is possible in the material. Cer- tain types of building, mechanically impossible in

1 In view of the fact that the word "organism," which implies the existence of a unity and order in the entity so designated, is universally accepted and employed, the word "organismic" is not only biologi- cally and etymologically justified but fills a need which is becoming more and more apparent. The word was used by Rhumbler some fifteen years ago (1905). Ritter (1919) has used the word "organismal" with a somewhat different connotation. wooden material, become possible when the materials are steel or concrete. Protoplasm in consequence of its physicochemical constitution represents a certain kind of material in which certain kinds of pattern are possible, and organismic pattern must lie within the range of possibilities determined by this constitution.

As regards the most minute and presumably the simplest organisms, e.g., certain bacteria and ultra- microscopic forms, if such exist, our knowledge is at present insufficient to permit us even to say with cer- tainty whether they are cells in the ordinary sense of the word, and concerning their pattern as organisms we are almost completely ignorant, but it is at least certain that in the bacteria orderly differences between surface and interior exist. The single cell is of course primarily an organism, though it may as a part of a multicellular organism become specialized and lose to a greater or less degree its capacity for independent exist- ence. And, finally, the multicellular form represents a still more complex order and integration.

The fundamental fact for present purposes is that the organism, as the term implies, represents an order and unity of some sort. Concerning the character of this order in the simplest living forms we know practi- cally nothing, but as regards all other organisms we may say that this order expresses itself in the appearance of more or less definite spatial morphological and functional relations and in a more or less definite and orderly behavior of the whole. It is necessary before going farther to inquire how the organismic pattern or order

differs from the non-organismic. In the first place the organismic pattern is fundamentally a molar, not a molecular or atomic pattern, for it involves regions con- sisting of many molecules usually of many different kinds. The pattern of protoplasm is apparently submolecular, molecular, and colloidal, but the organism involves some- thing more, viz., an " organization'1' of the protoplasm into different molar regions of different constitution or structure. Since these different regions or organs are dynamically active and yet the organism is an orderly whole, physicochemical relations of some sort must exist between the different regions. These relations again are not interatomic or intermolecular, like the relations of ordinary chemical reactions, nor are they like the relations between colloid particles. They are inter-regional relations involving distances of much greater order of magnitude than any of the relations in purely protoplasmic pattern, so that they apparently involve some sort of action at a distance or some means of communication.

In short the organismic pattern is of a very different order of magnitude from the protoplasmic pattern, involving molar dimensions far beyond those of any protoplasmic constituent. It is a differentiation and integration of different regions of protoplasm or of different cells or cell groups and it evidently cannot be interpreted in the same terms as protoplasmic pattern. It may exist within the limits of a single cell or it may involve millions of cells: it may be evanescent, tem- porary, or readily modified by change in external conditions, or it may be persistent to a high degree, but in all cases it represents a pattern of a different order of

magnitude from the protoplasmic pattern, and there- fore involves integrating and ordering factors which must be different, at least in degree, from those con- cerned in protoplasmic integration. Manifestly an adequate conception of the nature of organismic integration and pattern is essential to the progress of biology, because all except the most super- ficial knowledge of the organism as a whole must be based upon such a conception. Without it the data of biological research have a significance in relation to the living organism somewhat like that of the house- wrecker's stock in trade to the house.

In order to find a point of departure for further attack on the problem it is necessary to consider first whether organismic pattern is inherent in protoplasm or imposed upon it from without. In other words, does organismic pattern develop, so to speak, spontaneously out of protoplasm or the cell, or is it in some sense a response to environmental factors ? Any extended historical consideration of the problem of the origin of pattern or integration in organisms is impossible at this time, but it may be pointed out that the preformists, and notably among them Weismann, have simply assumed as inherent in protoplasm such integration and pattern as seemed to be necessary and have described it in terms of hypothetical entities of some sort. Essentially the same point of view persists among those who postulate a definite, persistent mor- phological pattern in the chromosome (Morgan, 1919, chap. xix). Preformistic theory offers us nothing but

description and statement of problems in certain terms; moreover, the implications of such theory are vitalistic rather than mechanistic. On the other hand, the epigenetic school in attempt- ing to derive organismic pattern from the relations between living protoplasm and its environment has found it difficult to account for the definiteness and constancy of the organismic unity and order and has failed thus far to formulate any adequate conception of the epigenetic origin and development of organismic pattern. And finally, the vitalist recognizes and states the problem clearly, but attempts to solve it by postu- lating the existence of a non-mechanistic integrating and ordering principle peculiar to the organism. That is to say, he merely substitutes a metaphysical problem for the scientific problem. In view of the fact that mechanistic possibilities as regards patterns in organisms are by no means exhausted the vitalistic assumptions are not at present justified as anything more than matters of opinion. Preformistic theory can do no more than state the problem in certain terms, and epi- genetic theory has not progressed very far toward a solution. In view of this situation it seems necessary as the first step in an attack upon the problem to under- take some further analysis in the attempt at a new formulation.

The organism represents an order and unity in protoplasm which is related at every point to the external world. The development and evolution of organismic integration are essentially the evolution of mechanisms and methods of response and adjustment to environmental conditions. If organismic pattern were inherent in protoplasm and a direct result of protoplasmic constitution we should expect it to show a close relation to the differences in constitution, e.g., some such relation as that which exists between constitution and crystalline form in the different haemoglobins (Reichert and Brown, 1909). While it is true that no two species or organisms and even no two individuals are alike in all details, yet the more general features of organismic pattern are apparently to a large extent independent of specific differences in protoplasmic constitution. As regards the general plan of spatial arrangement of parts and developmental sequence we find three general types of organismic pattern of very wide occurrence, the com- pletely radiate, in which the pattern is referable to a center, the axiate or polar, in which it is referable to a line, and the bilateral, in which it is referable to a plane. At least most organisms show some one or some com- bination of more than one of these patterns. Physio- logically the most general feature of organismic pattern appears to be that of dominance and subordination, of controlling and being controlled.

The attempts to discover terms in which a pre- formistic conception of organismic pattern might be formulated have been largely speculative and without satisfactory results. For example, the conception of organismic pattern in purely molecular terms, i.e., of the organism as crystalloid, either as a whole or as a con- geries of minute crystals similarly oriented, not only fails to account for the facts, but finds its basis in remote and apparently very superficial analogies, and certain other suggestions along stereochemical lines are purely speculative. The assumption of some other undetermined

basis of arrangement and order resulting from the inter- action of the hypothetical protoplasmic entities, deter- minants, pangenes, or whatever we choose to call them, is not only more extremely speculative but fails completely as a mechanistic hypothesis. To account adequately in any such terms for the uniformity, defi- niteness, and constancy of organismic pattern requires the further postulation of something very closely resem- bling the ordering and integrating principle of the vitalists. And finally, even though we invoke the aid of variation and natural selection at every step, as Weismann has done, our increasing knowledge of the facts of adaptation and of the amazing variety and delicacy of regulatory response to environmental con- ditions present always new difficulties to the pre- formistic interpretation.

Turning to the other mechanistic alternative, the imposition of organismic pattern upon protoplasm through its relations with the external world, it is clearly necessary for any adequate conception first, that we know something of the fundamental nature of organismic pattern, and second, that we determine, experimentally if possible, how such pattern originates in the relations between protoplasm and its environment. One of the most striking aspects of organismic pattern, to which attention has already been called (see p. 8), viz., the similarity as regards various of the more general features in different protoplasms indicates that such pattern, at least as regards these features, cannot be very closely associated with the specific differences in the proto- plasms of different species or groups and suggests the possibility that it may be primarily quantitative in

character, that is to say, that such similarities of pattern may result from the existence of similar quantitative situations and relations in protoplasms which are specifically very different in constitution. But in spite of the fact that many different lines of evidence point toward such a conception, and although many authors have called attention to the quantitative character of various aspects of organismic pattern, the distinction between pattern and material and between the quanti- tative and the specific or qualitative aspects of proto- plasmic and cellular behavior has often not been clearly drawn. Particularly is this true of late years, during which chemical correlation, internal secretions, hormones, i.e., the production, transportation and effect in the organism of various substances, supposed or known to be specific in their constitution and action, has received so much attention that certain other aspects of life have been to a large extent ignored, as regards their relation to integration or pattern in the organism. Nevertheless we must at least raise the question, To what extent is physiological integration a matter of quantitative relations in protoplasms of dif- ferent specific constitution ?