7* * «?>r V ; UNIVERSITY OF ftlrtlf OO l LIBRARY . ' ' CALIFORNIA * NOV23 1891 , /< OL m ssions No. Shelf No. ARY A TEXT BOOK OP PHYSIOLOGY A TEXT BOOK OF PHYSIOLOGY BY M. FOSTER, M.A., M.D., LL.D., F.E.S., PROFESSOR OF PHYSIOLOGY IN THE UNIVERSITY OF CAMBRIDGE, AND FELLOW OF TRINITY COLLEGE, CAiMBRIDGE. WITH ILLUSTRATIONS. FIFTH EDITION, LARGELY REVISED. PART III. The Central Nervous System. Uon&on : MACMILLAN AND CO. AND NEW YORK. 1890 [The Right of Translation is reserved.} BIOLOQY UBRARY G PRINTED BY C. J. CLAY, M.A. AND SONS, AT THE UNIVERSITY PRESS. First Edition 1876. Second Edition 1877. Third Edition 1879. Fowr«/i jBdt*ton 1883. Reprinted 1884, 1886. Jft/tfc Edition, 1890. T AM of course aware of the disadvantages of issuing this edition of my Text Book in instalments, and very much regret that this part does not complete the work, The failure to get the whole of the remainder ready has been dive to lack, not of will, but of ability and opportunity. I take this opportunity of thanking my friend Dr Gowers, for the loan of two woodcuts, as well as for much valuable advice. Throughout the whole of this part I have been largely assisted by my colleague Mr Langley, and by my friend and former pupil Dr Sherrington. The latter, besides helping me with criticisms, has prepared for me most of the figures after original drawings by himself. What little merit there may be in this part is largely due to these two gentlemen. M. FOSTER CAMBRIDGE, September, 1890. CONTENTS OF PART III. BOOK III. THE CENTBAL NERVOUS SYSTEM AND ITS INSTRUMENTS. CHAPTER I. THE SPINAL CORD. SECTION I. ON SOME FEATURES OF THE SPINAL NERVES. PAGE § 558. The spinal nerves 849 § 559. On efferent and afferent impulses 850 § 560. Efferent fibres run in the anterior root and afferent fibres in the posterior root ..." 852 § 561. The "trophic" influence of the ganglion of the posterior root; the degeneration of nerve fibres 853 SECTION II. THE STRUCTURE OF THE SPINAL CORD. § 562. The general features of the cord ; grey and white matter . . 856 § 563. The structure of the white matter ; neuroglia 859 § 564. The structure of the grey matter 861 § 565. The central canal, the substantia gelatinosa centralis, and the substantia gelatinosa of Rolando 863 § 566. The grouping of the nerve cells. The cells of the anterior and posterior horn, the lateral group, Clarke's column, and the lateral horn. The reticular formation 865 § 567. The tracts of white matter. Median posterior column, external posterior column. The evidence of the differentiation of the white matter into tracts. Ascending and descending degeneration. Descending tracts: crossed and direct pyramidal tracts, antero- lateral descending tract. Ascending tracts: cerebellar tract, antero-lateral ascending tract, median posterior tract . . . 869 viii CONTENTS. PAGE §568. The meaning of the terms "ascending" and "descending" de- generation, and the inferences to be drawn from them . . . 875 § 569. The connections of the nerve roots ; of the anterior root ; of the posterior root, median, lateral and intermediate bundles . . 876 § 570. The special features of the several regions of the spinal cord. The conns medullaris, the lumbar and cervical swellings. Variations in the sectional area of the white matter 878- § 571. Variations in the sectional area of the grey matter .... 880 § 572. The relative size, form and features of transverse sections of the cord at different levels 881 § 573. Variations in the disposition of nerve cells and groups of nerve cells at different levels 88& § 574. Variations in the several columns of white matter at different levels. 886 § 575. The course of the crossed and of the direct pyramidal tract along the length of the cord 88S- § 576. The course of the cerebellar tract along the length of the cord . 890 § 577. The course of the median posterior tract along the length of the cord 891 § 578. The course of the antero-lateral ascending tract along the length of the cord 895 § 579. The nature of the grey matter of the cord ; the segmental ground work, the nerve cells 895 § 580. The nature and relation to the grey matter of the tracts of white matter 899 § 581. Longitudinal commissural tracts, and transverse connections . . 900> SECTION III. THE REFLEX ACTIONS OF THE CORD. § 582. The difficulties attending the experimental investigation of the central nervous system; 'shock' and other effects of an operation. 902 § 583. The differences, as regards reflex movements, between different kinds of animals 904 § 584. The features of a reflex act dependent on the character of the afferent impulses 905 § 585. The complex nature of the central processes in a reflex movement . 906 § 586. The characters of a reflex movement dependent on the strength of the stimulus . . ; . . . . . . . . 906 § 587. The characters of a reflex movement dependent on the part of the body to which the stimulus is applied 907 § 588. The complexity of many reflex movements ; their relation to intel- ligence 908 § 589. Eeflex movements coordinated by afferent impulses other than the exciting impulses ; relations to consciousness .... 910 § 590. The characters of a reflex movement determined by the intrinsic condition of the cord . . .'.*'. 912 § 591. The reflex movements carried out by the spinal cord in man . . 912 Reflex actions resulting in changes other than movements . . 914 The inhibition of reflex actions ." . .' . . . . 915 §594. The time required for reflex actions . "."'..' . , . . 918 CONTENTS. ix SECTION IV. THE AUTOMATIC ACTIONS OF THE SPINAL CORD. PAGE § 595. Automatic actions of the spinal cord in the frog and in the dog . 920 § 596. Automatic activity dependent on afferent impulses .... 921 § 597. Tone of skeletal muscles 922 § 598. Tendon phenomena, knee jerk 926 § 599. Rigidity of muscles through spinal action 927 CHAPTER II. THE BRAIN. SECTION I. ON SOME GENERAL FEATURES OF THE STRUCTURE OF THE BRAIN. § 600. The embryonic brain ; the three primary cerebral vesicles . . 929 § 601. The transformation of these into the bulb and fourth ventricle, the cerebellum and pons varolii, the cura cerebri, corpora quadri- gemina, and third ventricle 930 § 602. The vesicles of the cerebral hemispheres, their growth and trans- formation into the cerebrum ; the cerebral hemispheres, corpus striatum, corpus callosum, fornix, and choroid plexus . . . 930 § 603. The parts of the adult brain corresponding to the main divisions of the embryonic brain 935 § 604. The cranial nerves 936 SECTION II. THE BULB. § 605. The main changes by which the cervical spinal cord becomes trans- formed into the bulb ; the pyramids and their decussation, the olivary bodies, the fasciculus cuneatus and fasciculus gracilis . 937 § 606. The superior or sensory decussation 942 § 607. The opening up of the central canal of the spinal cord into the fourth ventricle of the bulb ; the calamus scriptorius . . . 943 § 608. The changes in the grey matter ; the reticular formation and the arcuate fibres 944 § 609. The olivary nucleus, or inferior olive, the inner and outer accessory olivary nuclei, the antero-lateral nucleus 945 § 610. The gracile and cuneate nuclei ; the changes in the gelatinous substance of Eolando 947 §611. The fibres of the bulb 948 § 612. The relations of the gracile and cuneate nuclei to the inter-olivary layer, to the fillet, and to the restiform body .... 949 CONTENTS. SECTION III. THE DISPOSITION AND CONNECTIONS OF THE GREY AND WHITE MATTER OF THE BRAIN. PAGE § 613. The chief collections of grey matter 952 1. The central grey matter and the nuclei of the cranial nerves. § 614. The central grey matter .......... 953 § 615. The nuclei of the cranial nerves, their topographical distribution . 953 § 616. The nucleus of the twelfth or hypoglossal nerve .... 954 § 617. The nuclei of the eleventh or spinal accessory, tenth or vagus, and ninth or glossopharyngeal nerves . -V . . . . . 955 § 618. The nuclei of the eighth or auditory nerve * . . . 957 §619. The nucleus of the seventh or facial nerve . . . • 960 § 620. The nucleus of the sixth or abducens nerve . . . . • 962 § 621. The nuclei of the fifth or trigeminal nerve 962 § 622. The nucleus of the fourth or trochlear nerve 965 The nucleus of the third or oculomotor nerve 965 On the nature and relations of the several nuclei of the cranial nerves . . . . 967 2. The superficial grey matter. § 625. The cortex of the cerebrum and the superficial grey matter of the cerebellum . . . ,.'.;.. . . 970 3. The intermediate grey matter of the Crural system. § 626. The corpus striatum and optic thalamus, their positions and relations. The crus composed of pes and tegmentum. The internal capsule and corona radiata 970 § 627. The nucleus lenticularis ; globus pallidus and putamen. The nucleus caudatus 974 The optic thalamus, its nuclei ; the pulvinar 978 The substantia nigra, the red nucleus, and the corpus subthalamicum. The grey matter of the pons, the .upper olive .... 981 4. Other collections of grey matter. § 630. The corpora quadrigeniina. The corpora geniculata. The corpus dentatum of the cerebellum 983 The arrangement of the Fibres of the Brain. § 631. The pedal and tegmental systems 984 Longitudinal Fibres of the Pedal System. § 632. The pyramidal tract. Anterior or frontal cortical fibres. Posterior or temporo-occipital fibres. Fibres from the nucleus caudatus to the crus 987 CONTENTS. xi Longitudinal Fibres of the Tegmental System. PAGE § 633. Cortical fibres, the optic radiation 991 § 634. The superior peduncles of the cerebellum. The fillet. The longitu- dinal posterior bundle. Tracts from the corpora quadrigemina . 992 Transverse or Commissural Fibres. § 635. The corpus callosum. The anterior white commissure. The fornix. The middle peduncles of the cerebellum 994 § 636. A summary of some of the chief relations of the several parts of the brain to each other and to the spinal cord 996 SECTION IV. ON THE PHENOMENA EXHIBITED BY AN ANIMAL DEPRIVED OF ITS CEREBRAL HEMISPHERES. § 637. The absence of distinct signs of volition and intelligence . . 999 § 638. The characters of the movements of a brainless frog . . . 1000 § 639. The phenomena exhibited by birds after removal of their cerebral hemispheres 1003 § 640. The effects of removing the cerebral hemispheres in mammals . 1005 § 641. The effects of removing the cerebral hemispheres in the dog . . 1006 SECTION V. THE MACHINERY OP COORDINATED MOVEMENTS. § 642. The effects of injury to the semicircular canals. Our appreciation of the position of our body, the sense of equilibrium . . . 1009 § 643. Afferent impulses and sensations as factors of the coordination of movements 1013 § 644. The phenomena and causation of vertigo 1015 § 645. Forced movements 1017 § 646. The parts of the middle brain concerned in the coordination of movements 1019 SECTION VI. ON SOME HISTOLOGICAL FEATURES OF THE BRAIN. § 647. The structure of the central grey matter and some other collections of grey matter 1021 § 648. The histology of the superficial grey matter of the cerebellum ; the structure of the nuclear layer, molecular layer and of the cells of Purkinje 1022 § 649. The cerebral cortex. The general features of the grey matter ; the pyramidal or other cells 1026 § 650. The layers of the grey matter 1029 § 651. The histological features of the parietal, occipital and frontal regions 1030 § 652. The probable significance of the structure of the cortex . . . 1032 xii CONTENTS. SECTION VII. ON VOLUNTARY MOVEMENTS. PAGE § 653. The real distinction between voluntary and involuntary movements. 1034 § 654. The cortical motor areas of the dog ; the characters of the move- ments resulting from cortical stimulation 1035 § 655. The cortical motor areas in the monkey 1038 § 656. The cortical motor areas in the anthropoid ape .... 1043 § 657. The movements of cortical origin carried out by means of the pyramidal tract ; the nature of the movements so carried out . 1044 § 658. The results of the removal of a cortical area in dog and in the monkey 1049 § 659. The cortical motor areas in man ; the area for speech . . . 1052 § 660. The nature of the action of a motor area in carrying out a voluntary movement ; the characters of aphasia 1056 § 661. The same as illustrated by the area for a limb in the dog ; the influence of sensory impulses 1058 § 662. The relations of the motor area to other parts of the central nervous system ; the motor area employed in movements usually called involuntary 1061 § 663. The passage of volitional impulses along the spinal cord in animals. 1063 § 664. Their passage in man 1065 § 665. A summary of the chief facts concerning the carrying out of voluntary movements . . 1066 SECTION VIII. ON THE DEVELOPMENT WITHIN THE CENTRAL NERVOUS SYSTEM OF VISUAL AND OP SOME OTHER SENSATIONS. § 066. Visual impulses and sensations ; visual fields, and binocular vision. 1070 § 667. The decussation of the optic nerves in the optic chiasma . . 1073 §668. The course of the optic tract . ...... . . 1074 § 669. The endings of the optic tract in the lateral corpus geniculatum, the pulvinar and the anterior corpus quadrigeminum ; the results of degeneration and atrophy experiments 1075 § 670. The connection of the three above bodies with the cerebral cortex ; the meaning of the terms, blindness total and complete or partial, hemianopsia, amblyopia. The difficulties of interpretation attend- ing experiments on the vision of animals 1076 § 671. The nature of the movements of the eyes caused by stimulation of the occipital cortex . . .-. , • , 1079 § 672. The effects on vision of removing parts of the occipital cortex in monkeys and in dogs ; the teachings of clinical histories . . 1081 § 673. The probable progressive development of visual sensations ; lower and higher visual centres 1083 § 674. Sensations of smell. The structure of the olfactory bulb and tract ; the connections of the tract with other parts of the cerebrum . 1085 CONTENTS. xiii PAGE 675. The cortical area for smell 1087 676. Sensations of taste 1087 677. Sensations of hearing 1088 SECTION IX. ON THE DEVELOPMENT OF CUTANEOUS AND SOME OTHER SENSATIONS. § 678. Sensations of touch, heat, cold and pain 1090 § 679. Theoretical difficulties touching the cortical localisation of cutaneous sensations. The effects on cutaneous sensations of removing regions of the cortex 1091 § 680. The afferent tracts from the spinal cord, their endings in the brain. 1094 § 681. The effect of sections of the spinal cord on the transmission of afferent impulses influencing the vasomotor centre . . . 1096 § 682. Other experiments on animals as to the effects of sections of the spinal cord on the transmission of sensory impulses . . . 1099 § 683. The teachings of clinical histories ; different paths for different sensory impulses . . 1101 § 684. General considerations on the development of sensations along the spinal cord. The cerebellar tract, the median posterior tract, the grey matter and internuncial tracts 1102 § 685. The terms 'sensory' and 'motor' not an adequate description of the processes in the central nervous system 1105 § 686. The transmission of sensations within the brain. The relations of the cerebellum 1106 SECTION X. ON SOME OTHER ASPECTS OP THE FUNCTIONS OP THE BRAIN. § 687. Considerations touching the cerebellum 1109 § 688. Considerations touching the corpora quadrigemina .... 1112 §689. The splanchnic functions of the brain . . • . . . 1114 § 690. General considerations on the processes taking place in the cortex. The sources of the energy of the cortex 1115 SECTION XI. ON THE TIME TAKEN UP BY CEREBRAL OPERATIONS. 691. The reaction period or reaction time 1120 692. Elementary analysis of psychical processes, the time taken up by each. The time required for discrimination, for the development of perception, and of the will ; the circumstances influencing them. 1122 xiv CONTENTS. SECTION XII. THE LYMPHATIC ARRANGEMENTS OF THE BRAIN AND SPINAL CORD. PAGE § 693. The membranes of the brain and spinal cord 1125 § 694. The sources of the cerebrospinal fluid 1126 § 695. The characters of the cerebrospinal fluid 1128 § 696. The renewal of the cerebrospinal fluid. The purposes served by the fluid 1129 SECTION XIII. THE VASCULAR ARRANGEMENTS OF THE BRAIN AND SPINAL CORD. § 697. The distribution and characters of the arteries of the brain . . 1131 § 698. The venous arrangements of the brain ; the venous sinuses . . 1133 § 699. The supply of blood to the brain relatively small. The methods of investigating the circulation of the brain 1134 § 700. The supply of blood to the brain modified by the respiration and by changes in the general arterial pressure. The want of clear proof of special vasomotor nerves for the cerebral arteries . . . 1136 § 701. The flow of blood through the brain nevertheless influenced by changes taking place in the brain itself 1138 LIST OF FIGURES IN PART III. FIG. PAGE 96. A transverse dorsoventral section of the spinal cord (human) at the level of the sixth thoracic nerve 857 97. Diagram to illustrate the nature of the substance of Kolando . . 864 98. Transverse dorsoventral section of the spinal cord (human) at the level of the sixth cervical nerve 866 99. Transverse dorsoventral section of the spinal cord (human) at the level of the third lumbar nerve . 868 100. Diagram to illustrate the general arrangement of the several tracts of white matter in the spinal cord 872 101. Diagram shewing the united sectional areas of the spinal nerves proceeding from below upwards 879 102. Diagram shewing the variations in the sectional area of the grey matter of the spinal cord, along its length 880 103. Diagram shewing the relative sectional areas. of the spinal nerves as they join the spinal cord 880 104. Diagram illustrating some of the features of the spinal cord at diffe- rent levels 882 105. Diagram shewing the variations in the sectional area of the lateral columns of the spinal cord, along its length 886 106. Diagram shewing the variations in the sectional area of the anterior columns of the spinal cord, along its length 886 107. Diagram shewing the variations in the sectional area of the posterior columns of the spinal cord, along its length 886 108. Outlines of parts of the brain ; A dorsal, B lateral, C ventral aspect 938 109. Transverse dorsoventral sections of the bulb at different levels . . 940 110. Transverse dorsoventral section through the bulb just behind the pons 948 111. Transverse dorsal section through the bulb at the widest part of the fourth ventricle 958 112. Transverse dorsoventral section through the pons at the exit of the fifth nerve 961 113. Transverse dorsoventral section through the fore part of the pons . 96S 114. Transverse dorsoventral section through the crus and anterior corpora quadrigemina 964 115. Diagram to illustrate the position of the nuclei of the cranial nerves . 96(5 116. Diagrammatic outline of a dorsoventral section through the right hemisphere, at a level just posterior to the knee of the internal capsule 973 xvi LIST OF FIGURES IN PART III. FIO. PAGE 117. Diagrammatic outline of a dorsoventral section through the right hemisphere at a level anterior to fig. 116 975 118. Diagrammatic outline of a transverse dorsoventral section through the right hemisphere through the frontal lobe .... 976 119. Diagrammatic outline of a sagittal section taken through the right hemisphere seen from the mesial surface 977 120. View of right half of brain, as disclosed by a longitudinal section in the median line through the longitudinal fissure .... 979 121. Outline of horizontal section of brain, to shew the internal capsule . 985 122. Outline of a sagittal section through the hemisphere .... 986 123. Outline of a transverse dorsoventral section of the right half of the brain 988 124. The areas of the cerebral convolutions of the dog .... 1036 125. Outline of brain of monkey to shew the principal sulci and gyri . 1040 126. Left hemisphere of the brain of monkey viewed from the left side and from above 1041 127. Mesial aspect of the left half of the brain of monkey .... 1042 128. Diagram to illustrate the relative size of the pyramidal tract in man, monkey and dog 1049 129. Diagram of the convolutions and fissures on the lateral surface of the right cerebral hemisphere of man 1054 130. The same on the mesial surface 1054 131. The right lateral aspect of the cerebrum of man in outline to illus- trate the cortical areas 1055 132. Mesial surface of the right cerebral hemisphere of man in outline to illustrate the cortical areas . . . . . . . 1055 133. Diagram to illustrate the nervous apparatus of vision in man . . 1072 BOOK III. THE CENTRAL NERVOUS SYSTEM AND ITS INSTRUMENTS. F. 54 CHAPTER I. THE SPINAL CORD. SEC. 1. ON SOME FEATURES OF THE SPINAL NERVES. § 558. WE have called the muscular and nervous tissues the master tissues of the body ; but a special part of the nervous system, that which we know as the central nervous system, the brain and spinal cord, is supreme among the nervous tissues and is master of the skeletal muscles as well as of the rest of the body. We have already (Book I. Chap. III.) touched on some of the general features of the nervous system, and have now to study in detail the working of the brain and spinal cord. We have to inquire what we know concerning the laws which regulate the discharge of efferent impulses from the brain or from the cord, and to learn how that discharge is determined on the one hand by intrinsic changes originating, apparently, in the substance of the brain or of the cord, and on the other hand by the nature and amount of the afferent impulses which reach them along afferent nerves. As we shall see the study of the spinal cord cannot be wholly separated from that of the brain, the two being very closely related. Nevertheless it will be of advantage to deal with the spinal cord by itself as far as we can. The medulla oblongata or spinal bulb1 we shall consider as part of the brain. But before we speak 1 The term medulla oblongata is not only long, but presents difficulties, since the word medulla is now rarely used to denote the whole spinal cord (medulla spinalis) but is generally used to denote the peculiar coat of a nerve fibre, the white substance of Schwann. In using instead the word bulb or if necessary, spinal bulb there is little fear of confusion with any other kind of bulb. The adjective is in not uncommon use, in such phrases as 'bulbar paralysis.' 54—2 850 SPINAL NERVES. [BOOK HI. of the spinal cord itself, it will be desirable to say a few words concerning the spinal nerves, that is to say the nerves which issue from the spinal cord. We have already seen (§ 96) that each of the spinal nerves arises by two roots, an anterior root attached to the ventral or anterior surface, and a posterior root attached to the dorsal or posterior surface of the cord. We have further seen that the latter bears a ganglion, a 'ganglion of the posterior root' or 'spinal ganglion/ and we have (§ 97) studied the structure of this ganglion. We stated at the same time that while the trunk of a spinal nerve contained both efferent and afferent fibres, the efferent fibres were gathered up into the anterior root and the afferent fibres into the posterior root ; but we gave no proof of this state- ment. § 559. Before we proceed to do so, it will be as well to say a few words on the terms 'efferent' and 'afferent.' By efferent nerve fibres we mean nerve fibres which in the body usually carry impulses from the central nervous system to peripheral organs. Most efferent nerve fibres carry impulses to muscles, striated or plain, and the impulses passing along them give rise to movements ; hence they are frequently spoken of as 'motor' fibres. But all efferent fibres do not end in or carry impulses to muscular fibres ; we have seen for instance that some efferent fibres are secretory. Moreover all the nerve fibres going to muscular fibres do not serve to produce movement ; some of them, as in the case of certain vagus fibres going to the heart, are inhibitory and may serve to stop movement. By 'afferent' nerve fibres we mean nerve fibres which in the body usually carry impulses from peripheral organs to the central nervous system. A very common effect of the arrival at the central nervous system of impulses passing along afferent fibres is that change in consciousness which we call a 'sensation'; hence afferent fibres or impulses are often called ' sensory ' fibres or impulses. But as we have already in part seen, and as we shall shortly see in greater detail, the central nervous system may be affected by afferent impulses, and that in several ways, quite apart from the development of any such change of consciousness as may be fairly called a sensation. We shall see reason for thinking that afferent impulses reaching the spinal cord, and indeed other parts of the central nervous system, may modify reflex or automatic or other activity without necessarily giving rise to a " sensation." Hence it is advisable to reserve the terms 'efferent' and 'afferent' as more general modes of expression than ' motor ' or ' sensory.' We have seen in treating of muscle and nerve, that the changes produced in the muscle serve as our best guide for determining the changes taking place in a motor nerve ; when a motor nerve is CHAP, i.] THE SPINAL CORD. 851 separated from its muscle (§ 72) the only change which we can appreciate in it is an electrical change. Similarly in the case of an afferent nerve, the central system is our chief teacher; in a bundle of afferent fibres isolated from the central nervous system, in a posterior root of a spinal nerve for instance, the only change which we can appreciate is an electrical change. To learn the characters of afferent impulses we must employ the central nervous system. But in this we meet with difficulties. In studying the phenomena of motor nerves we are greatly assisted by two facts. First, the muscular contraction by which we judge of what is going on in the nerve is a comparatively simple thing, one contraction differing from another only by such features as extent or amount, duration, frequency of repetition and the like, and all such differences are capable of exact measurement. Secondly, when we apply a stimulus directly to the nerve itself, the effects differ in degree only from those which result when the nerve is set in action by natural stimuli, such as the will. When we come, on the other hand, to investigate the phenomena of afferent nerves, our labours are for the time rendered heavier, but in the end more fruitful, by the following circumstances : — First, when we judge of what is going on in an afferent nerve by the effects which stimulation of the nerve produces in some central nervous organ, in the way of exciting or modifying reflex action, or modifying automatic action, or affecting consciousness, we are met on the very threshold of every inquiry by the difficulty of clearly distinguishing the events which belong exclusively to the afferent nerve from those which belong to the central organ. Secondly, the effects of applying a stimulus to the peripheral end- organ of an afferent nerve are very different from those of applying the same stimulus directly to the nerve-trunk. This may be shewn by the simple experience of comparing the sensation caused by bringing any sharp body into contact with a nerve laid bare in a wound with that caused by contact of an intact skin with the same body. These and like differences reveal to us a complexity of impulses, of which the phenomena of motor nerves gave us hardly a hint. We shall further see in detail later on that our consciousness may be affected in many different ways by afferent impulses ; we must distinguish not only sensory from other afferent impulses, but also different kinds of sensory impulses from each other. Certain afferent nerves are spoken of as nerves of special sense, and the nature of the afferent impulses passing along these special nerves together with the modifications of consciousness caused by arrival of these impulses at the central nervous system constitute by themselves a complex and difficult branch of study. In some of the problems connected with the central nervous system we shall have to appeal to the results of a study of these special 852 SPINAL NERVES. [BOOK in. senses; but, on the other hand, a knowledge of the central nervous system is necessary to a proper understanding of the special senses ; and on the whole it will be more convenient to study the former before the latter. We may, however, digress here to remark that the question whether an afferent impulse differs in itself from an efferent impulse is one of great difficulty. It is true that the electrical changes, which alone as we have said we can appreciate in an isolated piece of nerve, appear to be the same in both kinds of fibres; in each the electrical change is propagated in both directions and possesses the same features. But it would be hazardous to insist too much on this. Moreover, we must remember that what we call a nervous impulse, especially one provoked by artificial stimulation, constitutes a gross change in the nerve fibre, and that changes of a finer, more delicate nature, such as cannot be shewn by the coarse methods used to detect a ' nervous impulse,' may take place in, and be propagated along, a nerve fibre. We shall have occasion immediately to point out that the condition of an afferent nerve fibre along its whole length is dependent on a nerve cell in the ganglion of the posterior root ; the fibre when cut off from the nerve cell degenerates and dies. This means that in the intact fibre certain influences are propagated along the fibre from the cell in the ganglion to the peripheral endings of the fibre, that is to say in a direction the opposite of that taken by the ordinary afferent nervous impulses ; and it may be that in like manner in efferent fibres some influences are propagated centripetally from the peripheral endings to the central nervous system. Our knowledge of these influences is extremely limited ; but it is important to bear in mind the possibility of their occurrence. And we had this in view, when above, in speaking of efferent and afferent fibres, we used the phrase " usually carry impulses." § 560. The proof that the afferent and efferent fibres which are both present in the trunk of a spinal nerve are parted at the roots, the efferent fibres running exclusively in the ventral or anterior root and the afferent fibres exclusively in the dorsal or posterior root, is as follows. When the anterior root is divided, the muscles supplied by the nerve cease to be thrown into contractions either by the will, or by reflex action, while the structures to which the nerve is distributed retain their sensibility. During the section of the root, or when the proximal stump, that connected with the spinal cord, is stimu- lated, no sensory effects are produced. When the distal stump is stimulated, the muscles supplied by the nerve are thrown into contractions. When the posterior root is divided, the muscles supplied by the nerve continue to be thrown into action by an exercise of the will or as part of a reflex action, but the structures CHAP, i.] THE SPINAL CORD. 853 to which the nerve is distributed lose the sensibility which they previously possessed. During the section of the root, and when the proximal stump is stimulated, sensory effects are produced. When the distal stump is stimulated no movements are called forth. These facts demonstrate that sensory impulses pass exclusively by the posterior root from the peripheral to the central organs, and that motor impulses pass exclusively by the anterior root from the central to the peripheral organs; and as far as our knowledge toes the same holds good not only for sensory and motor but also >r afferent and efferent impulses. An exception must be made to the above general statement, on account of the so-called " recurrent sensibility " which is witnessed in conscious mammals, under certain circumstances. It some- times happens that when the distal stump of the divided anterior root is stimulated, signs of pain are witnessed. These are not caused by the concurrent muscular contractions or cramp which the stimulation occasions, for they persist after the whole trunk of the nerve has been divided some little way below the union of the roots above the origins of the muscular branches, so that no contractions take place. They disappear when the posterior root is subse- quently divided, and they are not seen if the mixed nerve-trunk be divided close to the union of the roots. The phenomena are probably due to the fact, that bundles of sensory fibres of the posterior root after running a short distance down the mixed trunk turn back and run upwards in the anterior root, (being distributed probably to the pia mater) and by this recurrent course give rise to the recurrent sensibility. § 561. Concerning the ganglion on the posterior root, we may say definitely that we have no evidence that it can act as a centre of reflex action ; nor have we any evidence that it can spontaneously give origin to efferent impulses and thus act as an automatic centre, as can the central nervous system itself. The bodies of the nerve cells behave somewhat differently from the axis cylinders at some distance from the cells, though, as we have seen, these are in reality processes of the nerve cells ; thus the nerve cells in the ganglion appear to be more sensitive to certain poisons than are the nerve fibres of the nerve-trunk. But beyond this, our know- ledge concerning the function of the ganglion is almost limited to the fact that it is in some way intimately connected with the nutrition of the nerve. As we have already (§ 83) said, when a mixed nerve- trunk is divided the peripheral portion degenerates from the point of section downwards towards the periphery. The central portion does not so degenerate, and if the length of nerve removed be not too great, the central portion may grow downwards along the course of the degenerating peripheral portion, and thus regenerate the nerve. This degeneration is observed when the mixed trunk is divided in any part of its course from the periphery 854 SPINAL NERVES. [BOOK in. to close up to the ganglion. When the posterior root is divided between the ganglion and the spinal cord, the portion attached to the spinal cord degenerates, but that attached to the ganglion remains intact. When the anterior root is divided, the proximal portion in connection with the spinal cord remains intact, but the distal portion between the section and the junction with the other root degenerates ; and in the mixed nerve-trunk many degenerated fibres are seen, which, if they be carefully traced out, are found to be motor (efferent) fibres. If the posterior root be divided carefully between the ganglion and the junction with the anterior root, the small portion of the posterior root left attached to the peripheral side of the ganglion above the section remains intact, as does also the rest of the root from the ganglion to the spinal cord, but in the mixed nerve- trunk are seen numerous degenerated fibres, which when examined are found to have the distribution of sensory (afferent) fibres. Lastly, if the posterior ganglion be excised, the whole posterior root degenerates, as do also the sensory (afferent) fibres of the mixed nerve-trunk. Putting all these facts together, it would seem that the growth of the efferent and afferent fibres takes place in opposite directions, and starts from different nutritive or 'trophic' centres. The afferent fibres grow away from the ganglion either towards the periphery, or towards the spinal cord. The efferent fibres grow outwards from the spinal cord towards the periphery. This difference in their mode of nutrition is frequently of great help in investigating the relative distribution of efferent and afferent fibres. When a posterior root is cut beyond the ganglion, or the ganglion excised, all the afferent nerves degenerate, and in the mixed nerve-branches these afferent fibres, by their altered condition, can readily be traced. Con- versely, when the anterior roots are cut, the efferent fibres alone degenerate, and can be similarly recognized in a mixed nerve-tract. When the anterior root is divided some few fibres in it do not, like the rest, degenerate, and when the posterior root is divided, a few fibres in the anterior root are seen to degenerate like those of the posterior root ; these appear to be the fibres which give to the anterior root its "recurrent sensibility." In the case of certain spinal nerves at all events, it has also been ascertained that when the posterior root is divided, while most of the fibres in the part of the root thus cut off from the ganglion but left attached to the cord degenerate, some few do not. These few appear to have their trophic centre not in the ganglion, but in some part of the spinal cord itself; we shall refer to these later on. This method of distinguishing nerve fibres by the features of their degeneration, called the "degeneration method," or sometimes from the name of the physiologist who introduced CHAP, i.] THE SPINAL CORD. 855 it, the " Wallerian method," has proved of great utility. Thus in the vagus nerve which is composed not only of fibres which spring from the real vagus root but also of fibres proceeding from the spinal accessory roots, the two may be distinguished by section of the vagus and spinal accessory roots respectively. We shall presently see that this method may be applied to the differentiation of tracts of fibres in the brain and spinal cord. SEC. 2. THE STRUCTURE OF THE SPINAL CORD. § 562. Lying within the vertebral canal the spinal cord is protected by its 'membranes/ the dura mater, the arachnoid membrane and the pia mater. The consideration of the arrange- ment of these membranes and of the structure of the dura mater and arachnoid we will leave until we come to speak of the vascular and lymphatic supplies of the central nervous system ; the histo- logy of the pia mater may more fitly come with that of the spinal cord itself. Along its whole length from its junction with the bulb to its termination in the filum terminate the spinal cord, while possessing certain general features, is continually changing as to special features. It will be convenient to study first the general structure of some particular part, for instance the middle of the thoracic (dorsal)1 region, and afterwards to point out the special features which obtain in the several regions. A transverse vertical section of either a fresh or a hardened and prepared spinal cord at the thoracic region possesses an outline which is roughly speaking circular. In the middle of the anterior or ventral surface is a vertical fissure, the ventral or anterior fissure (Fig. 96, A. F.) running some way across the thickness of the cord from the ventral towards the dorsal surface. Opposite to it on the posterior or dorsal surface is a corresponding, deeper but narrower, dorsal or posterior fissure (Fig. 96, P. F.) which, however, as we shall see, differs materially in nature from the 1 It is very desirable to use the terms 'dorsal' and 'ventral' for the parts of the cerebro-spinal axis which lie respectively near the dorsal or back part, and the ventral or belly part of the body, instead of the terms posterior and anterior; but if this is done, the use of the word dorsal to denote the region of the cord between the lumbar and cervical regions is apt to lead to confusion ; hence the introduction of the word thoracic. If this use of dorsal and ventral be adhered to, before and behind, above and below, may conveniently be used to denote nearer the head and nearer the tail (or coccyx) respectively; anterior and posterior may also be used in the same sense except in the case of anterior and posterior fissure and horn, which terms seem too much honoured by time to be thrown aside. CHAP, i.] THE SPINAL CORD. 857 P.F FIG. 96. A TKANSVEKSE DOKSOVENTRAL SECTION OF THE SPINAL CORD (HUMAN) AT THE LEVEL OF THE SIXTH THORACIC (DORSAL) NERVE. (Sherrington) 1. Magnified 15 times. One lateral half only is shewn. The large conspicuous nerve-cells (drawn from actual specimens) are shaded black to render their relative size, shape and position more obvious; the outline of the grey matter has been made thick and dark in order to render it conspicuous. A.F. anterior fissure. P.F. posterior fissure, c.c. central canal, c.g.s. central gelatinous substance. A.r. anterior root, P.r. lateral (or intermediate) bundle, P.r'. median bundle of posterior root of spinal nerve, p', p" fibres of posterior root passing p', indirectly through the substance of Eolando, p", directly into grey matter, a.g.c. anterior grey commissure, p.g.c. posterior grey com- missure, a.c. anterior white commissure, ant. col. anterior column, lat. col. lateral column, post. col. posterior column, s.g. the substance of Eolando. s. septum marking out the external posterior column or column of Burdach, e.p. from the median posterior column or column of Goll, m.p. 1. cells of the anterior horn. 3. posterior vesicular column or vesicular cylinder, or column of Clarke ; the area of the cylinder is defined by a dotted line. 4. cells of the intermedio-lateral tract or lateral horn. 6. cells of the posterior horn. 7. cells of the anterior cervix, y a tract of fibres passing from the vesicular cylinder to the lateral column. 1 For this and many succeeding figures I am deeply indebted to my friend and former pupil Dr Sherrington who has kindly prepared the figures for me from his original drawings. 858 STRUCTURE OF SPINAL CORD. [BOOK HI. anterior fissure, and ought to be called a septum rather than a fissure. Between the two fissures the substance of the cord is reduced to a narrow isthmus uniting the two lateral halves, which in a normal cord are like each other in every respect. In the middle of the isthmus lies the section of a small canal, the central canal (Fig. 96, c. c.), which is all that remains of the relatively wide neural canal of the embryo. Each lateral half consists of an outer zone of white matter surrounding, except at the isthmus, an inner more or less crescentic, or comma shaped mass of grey matter. The convexity of each crescent is turned towards the median line of the cord, the two crescents being placed back and back and joined together by the isthmus just spoken of. The somewhat broader anterior extremity of the crescent, or head of the comma, is called the anterior cornu or Aorn; and the narrower posterior extremity of the crescent, or tail of the comma, is called the posterior cornu or horn. The part by which each horn is joined on to the middle part of the crescent is called the cervix, anterior and posterior respectively. The isthmus joining the backs of the two crescents, like the crescents themselves, consists, for the most part, of grey matter, the band running posterior or dorsal to the central canal being called the posterior grey commissure (Fig. 96, p. g. c), and the band running anterior or ventral to the canal being called the anterior grey commissure (Fig. 96, a. g. c). The posterior fissure touches the posterior grey commissure, but the anterior grey commissure is separated from the bottom of the anterior fissure by a band of white matter, called the anterior white commissure or, more simply, the white commissure or sometimes the anterior commissure (Fig. 96, a. c). If the section be taken at the level of the origin of a pair of spinal nerves, it will be seen that the anterior or ventral root, piercing the white matter opposite the head of the comma in several distinct bundles (Fig. 96, A.r), plunges into the anterior cornu, while the posterior or dorsal root (Fig. 96, P.r, P./), having the appearance of a single undivided bundle, passes, in part at least, into the posterior horn. Both roots are dispersed length- ways along the cord, the hinder roots of one nerve being close to the foremost roots of the nerve below, but it is only the anterior roots which are dispersed sideways. The compact bundle of the posterior root divides, with tolerable sharpness, the white matter in each lateral half of the cord into a posterior portion lying between the posterior fissure and the posterior root (Fig. 96 post, col.}, which portion since, as we shall see, 1^ runs, in the form of a column along the length of the cord, is called the posterior column, and into a portion lying to the outside of the posterior root between it and the anterior fissure, called the antero- lateral column. This latter may be considered as further divided, by the entrance of the anterior roots into a lateral column (Fig. 96, CHAP, i.] THE SPINAL CORD. 859 lat. col) between the posterior root and the most external bundle of the anterior root, and into an anterior column (Fig. 96, ant. col.) between the anterior fissure and the most external bundle of the anterior root. The part traversed by the bundles of the anterior root, as they make for the anterior horn, accordingly belongs to the anterior column ; but some writers speak of the anterior column as lying between the anterior fissure and the nearest bundle of the anterior root, thus making the region of the anterior root belong to neither anterior nor lateral column. And indeed the distinction between the anterior and the lateral column is to a great extent an artificial distinction. § 563. The ' white matter ' consists exclusively of medullated fibres supported partly by connective tissue and partly by a peculiar tissue known as neuroglia, of which we shall presently speak. The fibres are of various sizes, .but many of them are large, and in all of them the medulla is conspicuous. They run for the most part longitudinally, so that in transverse sections of the cord nearly the whole of the white matter appears under the microscope to be composed of minute circles, the transverse sections of the lon- gitudinally disposed fibres, imbedded in the supporting structures. The 'grey matter' also contains medullated fibres, but these are for the most part exceedingly fine fibres possessing a medulla which appears to differ from that of an ordinary nerve-fibre, since it does not stain readily with osmic acid, but is rendered visible by special modes of preparation such as that known as Weigert's. Hence these fine fibres are not apparent in ordinary carmine or other specimens, and indeed their presence was for a long time overlooked. Besides these fine medullated fibres, if we may call them such, the grey matter contains, what the white matter does not, nerve-cells with branching processes, naked axis-cylinders, and delicate filaments arising from the division of axis-cylinders or from the branching of nerve-cells, all these various structures being imbedded in neuroglia. Owing to the relative abundance of the white refractive medulla, the white matter pojgesses in fresh specimens a characteristic opaque white colour; Hence the name. The grey matter from the relative scantiness of medulla has no such opaque whiteness, is much more translucent, and in fresh specimens has a grey or rather pinky grey colour, the reddish tint being due to the presence partly of pigment and partly of blood, for the blood vessels are much more abundant in the grey matter than in the white. The pia mater which closely invests the cord all round consists of connective tissue, fairly rich in elastic elements and abun- dantly supplied with blood vessels ; it is indeed essentially a vascular membrane and furnishes the nervous elements of the cord with their chief supply of blood. It sends in at intervals partitions or septa of the same nature as itself radiating towards the central grey matter. The narrow posterior fissure is com- 860 STRUCTURE OF SPINAL CORD. [BOOK in. pletely filled up by a large septum of this kind, indeed as we have said is in reality not a fissure but a large septum ; but the anterior fissure is too wide for such an arrangement ; the whole membrane dips down into this fissure, following the surface of the cord and being reflected at the bottom. From these primary septa, secondary finer septa still composed of ordinary fibrillated connective tissue, carrying blood vessels, branch off; but these are soon merged into the peculiar supporting tissue called, as we have said, neuroglia. This consists in the first place of small branching cells, lying in various planes. The branching is excessive, so that the body of the cell is reduced to very small dimensions, indeed at times almost obliterated, the nucleus disappearing while the numerous branches are continued as long fine filaments or fibres pursuing a devious but for the most part a longitudinal course. In the second place these cells and fibres or filaments are im- bedded in a homogeneous ground substance. Relatively to the fibres and ground substance the bodies of the cells (which are called Deiter's cells), especially bodies such as bear obvious nuclei, are very scanty; hence in sections, especially in transverse sections, of the cord the neuroglia has often a dotted or punctated appearance, the dots being the transverse sections of the fine lon- gitudinally disposed fibres imbedded in the ground substance. Examined chemically the neuroglia is found to be composed not like connective tissue of gelatine, but of a substance which appears to be closely allied to keratin, the chief constituent of horny epidermis, hairs and the like, § 435, and which has therefore been called neurokeratin, (see also § 68). And indeed this neuroglia, though like connective tissue a supporting structure, is not, like connective tissue, of mesoblastic, but of epiblastic origin. The walls of the neural canal of the embryo which are transformed into the spinal cord of the adult consist at first of epithelial, epiblastic cells; and while some of these cells become nervous elements, others become neuroglia. The epithelial cells which are destined to form neuroglia become exceedingly branched, while their originally protoplasmic cell-substance becomes transformed to a large extent into neurokeratin. The neuroglia fills up the spaces between the radiating larger septal prolongations of the pia mater and the finer branched septa which starting from the larger ones carry minute blood vessels into the interior of the white matter. In these spaces it is so arranged as to form delicate tubular canals, of very variable size, running for the most part in a longitudinal direction. Each of these tubular canals is occupied by and wholly filled up with a medullated nerve-fibre of corresponding size. A medullated nerve-fibre of the white matter of the spinal cord resembles a medullated nerve- fibre of a nerve (§ 68) in being composed of an axis-cylinder and a medulla; but it possesses no primitive sheath or neurilemma. This is absent and indeed is not wanted ; the tubular sheath of CHAP, i.] THE SPINAL CORD. 861 neuroglia affords in the spinal cord (and as we shall see in the central nervous system generally) the support which in a nerve is afforded by the neurilemma. Nodes are, according to most authors, absent, but some say they are present. The white matter of the cord consists then of a more or less solid mass of neuroglia, having the structure just described, which is permeated by minute canals, some exceedingly fine and carrying very fine 2/Lt fibres, others larger and carrying fibres up to the size of 15//,. This mass is further broken up into areas by the smaller and larger vascular connective tissue septa with the edges and endings of which the neuroglia is continuous. Most of the nerve- fibres, as we have said, run longitudinally and in a transverse section of the cord are cut transversely; but as we shall see fibres are continually passing into and out of the white matter, and in so doing take a more or less transverse course ; these however are few compared with those which run in a longitudinal direction. On the outside of the cord below the pia mater the neuroglia is developed into a layer of some thickness from which nerve fibres are absent ; this is often spoken of as an inner layer of the pia mater ; but being neuroglia and not connective tissue is of a different nature from the pia mater proper. A layer of this superficial neuroglia also accompanies the larger septa, and a considerable quantity is present in the large septum called the posterior fissure. The pia mater carries not only blood vessels but also lymphatics ; of these however we shall speak when we come to deal with the vascular arrangements of the whole of the central nervous system. § 564. In the grey matter we may distinguish the larger, more conspicuous nerve-cells and the rest of the grey matter in which these cells lie. We have already (§ 99) described the general features of these larger nerve-cells, and shall have pre- sently to speak of their special characters and grouping. Meanwhile the most important point to remember about them besides the fact that they vary largely in form and size is that while one process may or does become an axis cylinder of a nerve-fibre, the others rapidly branch, and breaking up into fine nerve filaments are lost to view in the rest of the grey matter. These larger nerve-cells form, however, a part only, and in most regions of the cord the smaller part, of the whole grey matter. In a transverse section from the thoracic region (Fig. 96) a few only of these larger nerve-cells are seen in the whole section, and though they appear more numerous in sections from the cervical and especially from the lumbar regions (Figs. 98, 99), yet in all cases they occupy the smaller part of the area of the grey matter. The larger part of the grey matter consists, besides a neuroglia supporting the nervous elements, of nerve filaments running in various directions and forming, not a plexus properly so called, but an interlacement of extreme complexity. These filaments are, on 862 STRUCTURE OF GREY MATTER. [BOOK HI. the one hand, the fine medullated fibres spoken of above as being recognized with difficulty, and, on the other hand, non-medullated filaments ranging from fairly wide and conspicuous naked axis- cylinders down to fibrils of extreme tenuity, the latter arising apparently either from the division of axis-cylinders of nerve fibres passing into or out of the grey matter or from the continued branching of processes' of nerve cells. By* the modes of prepara- tion now available it has been shewn that the fine medullated fibres so far from being rare, are in certain parts of the grey matter so abundant as even to preponderate over the non-medullated fibres or fibrils. Lastly, besides the conspicuous nerve cells spoken of above, which, though of various sizes, may all perhaps be spoken of as large, a very large number of other cells of small size, some of which at all events must be regarded as true nerve-cells, are present in the grey matter. The neuroglia in which all these structures, nerve-cells, fine medullated nerve-fibres, naked axis-cylinders and fine filaments, are imbedded is identical in its general characters with that of the white matter, but, as naturally follows from the nature of the nervous elements which it supports, is differently arranged. In- stead of forming a system of tubular channels it takes on the form of a sponge work with large spaces for the larger nerve-cells and fine passages for the nervous filaments. At the junction of the grey matter with the white matter, the neuroglia of the one is continuous with that of the other, and the connective tissue septa of the latter run right into the former ; the outline of the grey matter is not smooth and even, but broken by tooth-like processes due to the septa. Since, as we have just said, some of the true nerve-cells are very small, and since the nerve filaments like the neuroglia fibres are very fine and take like them an irregular course, it often becomes very difficult in a section to determine exactly which is neuroglia and which are nervous elements. The neuroglia cells may however be distinguished perhaps from the smaller nerve-cells by their nuclei not being so conspicuous or so relatively large as in a nerve-cell, and by their staining differently. The grey matter then may be broadly described as a bed of neuroglia, containing a certain number of branching nerve-cells, for the most part though not exclusively large and conspicuous, but chiefly occupied by what is not so much a plexus as an intricate interweaving of nerve filaments running apparently in all directions. Some of these filaments are fairly conspicuous naked axis-cylinders, and a few are easily recognized medullated fibres of ordinary size ; but by far the greater number are either exceedingly fine medullated fibres, whose medulla is only made evident by special modes of preparation or delicate fibrils devoid of medulla. With the nervous web formed by these filaments the branching processes of the nerve-cells, on the one hand, and the divisions of nerve-fibres passing into or out of the grey CHAP, i.] THE SPINAL CORD. 863 matter, on the other hand, appear to be continuous. It may be added that the grey matter is well supplied with blood vessels, these being in it, as stated above, relatively much more numerous than in the white matter. § 565. The central canal is lined by a single layer of columnar epithelial cells, which are generally described as bearing cilia ; but it is not certain that the processes which may be seen project- ing from the surfaces of the cells are really cilia. These epithelial cells rest not on a distinct basement membrane but on a bed of neuroglia, free apparently or nearly so from nervous elements, which surrounds the central canal and is sometimes spoken of as the substantia gelatinosa centralis (Fig. 96, c. g. s.). The attached bases of the epithelial cells are branched or taper to a filament, and become continuous with the branched cells or fibres of the neuroglia below. As we said above the neuroglia elements are transformed epithelial cells ; and the continuity of the cells, which retaining the characters of epitHelial cells form a lining to the canal, with the cells which have become branched and lost their epithelial characters indicates the epithelial origin of the latter. The central canal with the surrounding area of neuroglia forms the central part of the isthmus uniting the two lateral halves of the cord. Posterior (dorsal) to this central mass lies the posterior grey commissure (Figs. 96, 98, 99, p. g. c.), composed chiefly of fine filaments running transversely, and anterior (ventral) to it lies first the thinner anterior grey commissure (Figs. 96, 98, 99, a. g. c.) of a similar nature, and then the relatively thick white commissure (Figs. 96, 98, 99, a. c.) which is formed by medullated fibres crossing over from one side of the cord to the other, and thus constitutes a decussation of fibres along the whole length of the cord. On each side, the central mass of neuroglia of which we are speaking gradually merges into the central grey matter of the corresponding lateral half. The end or head (caput) as it is frequently called of the posterior horn is occupied not by ordinary grey matter, but by a peculiar tissue, the substantia gelatinosa of Rolando, which forms a sort of cap to the more ordinary grey matter but differs in size and shape in different regions of the cord. Cf. figs 96, 97, 98, s.g. In carmine and some other modes of preparation it is frequently stained more deeply than is the ordinary grey matter, and in such preparations is very conspicuous. It may be described as consisting of a somewhat peculiar neuroglia traversed by fibres of the posterior root, and containing a large number of cells, which, for the most part small, the cell-bodies being small relatively to the nuclei, are not all alike, some being probably nervous and others not. It takes origin from the cells forming the immediate walls of the embryonic medullary canal. In the embryo, this canal is relatively wide, though compressed from side to side, and F. 55 864 THE SUBSTANCE OF ROLANDO. [BOOK m. in transverse sections of the medullary tube appears at a certain stage as a narrow oval slit placed vertically, and reaching almost from the dorsal to the ventral surface. The dorsal part of this long slit is later on closed up by the coming together of the walls and the obliteration of the greater part of the cavity, leaving the ventral part to form a circular canal, which by the development of the anterior columns assumes the central position. During this closure of the dorsal part of the canal a mass of the cells lining the canal is cut from the rest on each side, and during the subse- quent growth takes up a position at the end of the posterior horn. Hence, though it never apparently contains any cavity, the sub- stance of Rolando may be regarded as an isolated portion of the walls of the medullary canal, which has undergone a development somewhat different from that of the portion which remains as the lining of the central canal. Traces of this origin may be seen even in the adult. Thus in the lower end of the cord, in what we shall speak of presently as the cbnus medullaris, the central canal widens out dorsally, and in section (Fig. 97, A) presents on each side a bay x, stretching out towards the position of the posterior horn. At this region of the cord, though both white and grey matter are developed on the ventral surface, the posterior columns do not meet on the dorsal surface, but leave the central canal covered only by tissue which perhaps may be called neuroglia, but ABC FIG. 97. DIAGRAM TO ILLUSTRATE THE NATURE OF THE SUBSTANCE OF KOLANDO. The figures are purely diagrammatic and are not drawn to the same scale. In all three figures the grey matter is shaded with fine lines and the white matter with dots. A. transverse section of the lower end of the conus medullaris in man. e. epithe- lium lining the medullary canal, x. lateral expansion of the canal. B. transverse section of the spinal cord of the calf in the lower thoracic region. r. substance of Bolando. c. central canal. G. transverse section through mid thoracic region of cord in man. is of peculiar nature and origin. In the calf, in a part of the dorsal region the substance of Rolando is not confined to the tip of the posterior horn, but is continued to meet its fellow in the middle line. Fig. 97, B. If we imagine the dorsal portion of the canal of A to be cut off from the ventral portion, its cavity to be obliterated, and the lining epithelium with some of the sur- rounding elements to undergo a special development, the condition CHAP, i.] THE SPINAL CORD. 865 in B is reached by the growth of the posterior columns. From B, the transition to the normal state of things as in 97, c, is a very slight one. The extreme dorsal tip of the horn being of a more open texture than the substance of Rolando, is sometimes called the zona spongiosa. § 566. The grouping of the nerve-cells. The nerve-cells, at all events the cells which are large enough to be easily and without doubt recognized to be nerve-cells, form, as we have seen, only a part of the grey matter, and in some parts of the cord, in the thoracic region for instance, are so sparse that in a section of the spinal cord in this region thin enough to shew its histological features satisfactorily, the bodies of a few only of such cells are visible (Fig. 96) ; the greater part of the grey matter consists not of the bodies of conspicuous nerve-cells, but of a mass of fibres and fibrils passing apparently in all directions. In the cervical (Fig. 98) and especially in the lumbar (Fig. 99) regions the nerve-cells are both absolutely and relatively more abundant ; but even in a section taken from the lumbar region the nerve- cells, all put together, form the smaller part of the whole area of grey matter. Moreover, in respect of the number of cells all the sections of even the same region of the cord are not alike. Seeing that the cord may be considered as growing out of the fusion of a series of paired ganglia, each ganglion corresponding to a nerve, cf. § 96, we may fairly expect to find the fusion not complete, so that the nerve-cells would appear more numerous opposite a nerve than in the middle between two nerves. In some of the lower animals this arrangement is most obvious, and there are some reasons for thinking that even in man the nerve-cells are metamerically increased at the level of each nerve. Even when casually observed it is obvious that the nerve-cells are not scattered in a wholly irregular manner throughout the grey matter, being for instance much more conspicuous in the anterior horn than elsewhere ; and more careful observation allows us to arrange them to a certain extent in groups. The cells of the anterior horn are for the most part large and conspicuous, 67 /* to 135 //- in diameter, branch out in various direc- tions, and present an irregular outline in sections taken in different planes. We have reason to think that every one of them possesses an axis-cylinder process which, in the case at all events of most of the cells, passing out of the grey matter becomes a fibre of the adjacent anterior root. They are obvious and conspicuous in all regions of the cord, though much more numerous and individually larger in the cervical and lumbar enlargements than in the thoracic region. We may further, with greater or less success, divide them into separate groups. In the cervical and lumbar regions a fairly distinct group of cells is seen lying on the median side of the grey matter close to the anterior column (Figs. 98, 99, 1). This may be called the 55—2 866 THE NERVE CELLS OF THE CORD. [BOOK in. Rr' Rr. FIG. 98. TRANSVERSE DORSOVENTRAL SECTION OF SPINAL CORD (HUMAN) AT THE LEVEL OF THE SIXTH CERVICAL NERVE. (Sherrington.) This is drawn on the same scale as Fig. 96, that is magnified 15 times. r. f. I. lateral reticular formation, r. f. p. posterior reticular formation, p'. fine fibres of lateral bundle of the posterior root ; p", p'" fibres of median bundle CHAP, i.] THE SPINAL CORD. 8G7 of posterior root, entering grey matter from external posterior column, x. grey matter of posterior horn. Sp. a. bundles of fibres belonging to the spinal accessory nerve; in the lateral reticular formation they are seen cut trans- versely, b. is a natural septum of connective-tissue marking out the cerebellar tract C. T. from the crossed pyramidal tract C, P. T. z. s. zona spongiosa. 2 a, j3, 7, lateral cells of the anterior horn. 5. Cells in the region of the lateral reticular formation. The other letters of reference are the same as in Fig. 96. median group. It appears also in the thoracic region (Fig. 96, 1) ; indeed the question arises whether all the cells of the anterior horn in this region do not belong to this group. The other cells so conspicuous in the lumbar and cervical enlargements, and therefore probably in some way associated with the limbs, may be spoken of as forming altogether a lateral group ; but we may, though with some uncertainty, subdivide them into two or three groups. Thus in the lumbar region a group of cells (Pig. 99, 27) lying near the lateral- margin of the more dorsal part or base of the horn may be distinguished, as a lateral subgroup, from the cells occupying the ventral lateral corner of the horn and forming a ventral or anterior subgroup (Fig. 99, 2a) ; and the same distinction, though with less success, may be made in the cervical region (Fig. 98). Further, we may perhaps in both regions distinguish a group of cells placed more in the very middle of the horn as a central subgroup (Figs. 98, 99, 2/3). But, in all cases, the separation of these cells, which we have spoken of as a whole as lateral cells, into minor groups, is far less distinct than the separation of the median group from these lateral cells, especially if we admit that in the thoracic region, the median group is alone clearly represented. In the thoracic region a group of rather smaller cells is seen at the base of the anterior horn, near to the junction with the isthmus (Fig. 96, 7). In the cervical and lumbar region these cells are very scanty (Figs. 98, 99, 7). The cells of the posterior horn contrast strongly with those of the anterior horn in being few, and for the most part small. They are branched ; and though we have reason to believe that, like the cells of the anterior horn, they possess each an axis-cylinder process, this is not easily determined by actual observation; the processes do not run out to join the posterior root as do the corre- sponding processes in the anterior horn and therefore are not so readily seen. These cells occur in all regions of the cord, and appear to be arranged into two more groups. The lateral margin of the posterior horn, at about the middle or neck of the horn, is along the whole length of the cord, but especially in the cervical region, much broken up by bundles of fibres passing in various directions and forming an open network, called the lateral reticular formation (Fig. 98, 99, r. f. lat). In all regions of the cord a number of cells are found associated with this reticular formation, forming the group of the lateral reticular formation (Figs. 98, 99, 5). In all regions of the cord also a group of cells (Figs. 96, 98, 99, 6) 868 THE NERVE CELLS OF THE CORD. [BOOK in. is Found in that part of the horn where, a little ventral to the substance of Rolando, the uniform field of grey matter is broken up into a kind of network by a number of bundles of white fibres running in various directions. This network has also been called a r.ff, - J FIG. 99. TRANSVERSE DORSOVENTRAL SECTION OF THE SPINAL CORD (HUMAN) AT THE LEVEL OF THE THIRD LUMBAR NERVE. (Sherrington.) This is drawn to the same scale as Figs. 96, 97 and in the same way except that the outline of the grey matter is not exaggerated. Pr'. median, Pr. inter- mediate, Pr". lateral bundles of posterior roots. The region comprised under m.t. is the marginal zone or Lissauer's zone. The other letters of reference are the same as in 96 and 98. The three figures 96, 98, 99 are intended to illustrate the main differential features of the cervical, thoracic, and lumbar cord. /( v CHAP, i.] THE SPINAL CORD. 869 * vV v^ ^^ reticular formation, and has received the name of posterior reticular formation (Figs. 98, 99, r. f. p.) to distinguish it from the lateral reticular formation just mentioned ; the two however in some regions (see Fig. 96) join each other, and thus cut off a ventral portion of the posterior horn containing nerve-cells from a dorsal portion, x in Figs. 98, 99, in which no obvious or conspicuous nerve-cells are present. The groups of cells just mentioned with the restrictions and modifications spoken of occur along the whole length of the cord ; but the group of cells to which we must now call attention is almost confined to a special region of the cord, or at least is but feebly represented elsewhere. In the thoracic region, especially in the lower thoracic region (we shall return to the limits of the group later on) at the base of the posterior horn (Fig. 96, 3) just ventral to the curve formed by the posterior grey commissure as this bends dorsally to join the posterior horn, is seen on each side of the cord a conspicuous group of cells known as Clarke s column or the posterior vesicular column or vesicular cylinder. The cells composing this group, though varying in size at different levels, are rather large cells, and are for the most part fusiform, with their long axis placed lengthways along the cord, so that in transverse sections they often appear to have a rather small round body. They are surrounded by and as it were imbedded in a mass of fine fibres, the area of which is indicated by a dotted line in Fig. 96. Also conspicuous in the thoracic region is another group of cells lying on the outer side of the middle of the grey matter at about the junction of the anterior and posterior horns. This is known as the intermedio-lateral tract and is sometimes called the lateral horn (Fig. 96, 4). The cells composing it are some- what small spindleshaped cells with their long axis placed trans- versely. The group is conspicuous as we have said in the thoracic regions ; it may be recognized in the lumbar region (Fig. 98, 4), but in the cervical region becomes confused with the most dorsally placed or lateral subgroup of the anterior horn. We shall however have to return to these groups of cells when we come to speak of the differences between the several regions of the cord. § 567. The tracts of white matter. At first sight the white matter of the cord appears to be of uniform nature. We can use the nerve roots to delimitate the anterior, posterior and lateral columns, but we appear to have no criteria to distinguish parts in each column. In the cervical and upper thoracic regions of the cord, a septum (Fig. 96, s.) in the posterior column, somewhat more conspicuous than the other septa, has enabled anatomists to distinguish an inner median portion, the median posterior column, commonly called the postero-median column or column of Goll (Fig. 96, m. p.\ from an outer lateral portion, the external posterior column, commonly called the postero-external column or 870 THE TRACTS OF WHITE MATTER. [BOOK HI. column of Burdach (Fig. 96, e.p,\ the lateral part of which, nearer the grey matter, has, for reasons which we shall see later on, been called the posterior root-zone. But beyond this neither the irregular septa nor other features will enable us to distinguish one part of the white matter as different in nature from another. Nor have we better success when with the scalpel we attempt to unravel out the white matter into separate strands. Nevertheless we have convincing evidence that the white matter is arranged in strands, or tracts, or columns, which have different connections at their respective ends, which behave differently under different circumstances, which we have every reason to believe carry out different functions, but which cannot be separated by the scalpel because each of them is more or less mixed with fibres of a different nature and origin. The evidence for the existence of these tracts is twofold. One kind of evidence is embryological in nature. When a nerve fibre is being formed in the embryo, either in the spinal cord or elsewhere, the essential axis cylinder is formed first and the less essential medulla is formed later. Now when the develop- mental history of the spinal cord is studied it is found that, in the several regions of the cord, all the fibres of the white matter do not put on the medulla at the same time. On the contrary, in certain tracts, the medulla of the fibres makes its appearance early, in others later. By this method it becomes possible to distinguish certain tracts from others. • Another kind of evidence is supplied by facts relating to the degeneration of the fibres of the white matter. We have seen (§ 561) that the degeneration of a nerve fibre is the result of the separation of the fibre from its trophic centre, and that while the trophic centre of the afferent fibres is in the ganglion on the posterior root, that of the efferent fibres is in some part of the spinal cord. In the case of the efferent fibres the degeneration might be spoken of as descending from the spinal cord to the muscles or other peripheral organs. In the case of the afferent fibres of the trunk of the nerve, the degeneration is also one descending from the ganglion down to the skin or other peri- pheral organ. When however the section is carried through the posterior root of a spinal nerve, the degeneration takes place in the part of the nerve between the section and the spinal cord, it runs up from the section to and into the spinal cord, and may therefore be called an ascending degeneration. Thus we may say that when a nerve trunk or when a nerve root is cut completely across, all the fibres which are thereby separated from their trophic centres, degenerate. When the nerve trunk is divided all the fibres below the section undergo descending degeneration. If the anterior root be cut across, all the fibres of the root below the section undergo descending degeneration. If the posterior root be cut across, all the fibres of the root above the section undergo CHAP, i.] THE SPINAL CORD. 871 ascending degeneration with the exception of certain fibres which do not degenerate at all, and of which we shall speak later on. When the spinal cord is cut across, for instance in the dorsal region, all the fibres of the white matter do not degenerate either in the part of the cord above the section or in the part below. Some fibres, and indeed some tracts of fibres degenerate, and some do not. Further, some tracts degenerate in the cord above the section, and thus undergo what has been called an ascending degeneration ; other tracts degenerate in the cord below the section, and thus undergo what has been called a descending degeneration. These terms must however be used with caution. When a nerve trunk is cut across, the degeneration actually descends, in the sense that the progress of the degenerative changes may be traced downwards; they begin at the section and travel downwards at a -rate sufficiently slow to permit a difference being observed between the progress of degeneration at a spot near the section and that at one farther off. After section of or injury to the spinal cord, however, it is not possible to trace any such progress either upwards or downwards ; in the tracts both above and below the section or injury, degeneration either begins simultaneously along the whole length of the degenerating tract, or progresses along the tract so rapidly that no differences can be observed as far as the stage of de- generation is concerned between parts near to and those far from the section or injury. When, for. instance, the cord is divided in the cervical region, subsequent examination of the tracts of so-called descending degeneration shews that the de- generation is as far advanced in the lumbar region far away from the section as in the cervical region just below the section. Applied to the spinal cord, therefore, the term descending de- generation means simply degeneration below the seat of injury or disease, ascending degeneration means simply degeneration above the seat of injury or disease. We may add that the histological features of the degeneration of fibres in the spinal cord are not wholly identical with those of the degeneration of fibres in a nerve trunk. Thus, the neurilemma with its nuclei being absent from the fibres of the cord, no proliferation of nuclei takes place ; the axis- cylinder and medulla simply break up, are absorbed and disappear. Similar degenerations, ascending, or descending or both, are seen when the section is not carried right through the whole cord, but particular parts of the cord are cut through or simply injured. And similar degenerations occur as the consequences of disease set up in parts of the cord. In this way the results of sections of or of other injuries to or of diseases of the spinal cord have enabled us to mark out certain tracts of the white matter as undergoing degeneration and others as not, and moreover certain tracts as undergoing descending and 872 THE TRACTS OF WHITE MATTER. [BOOK HI. others as undergoing ascending degeneration. Further, the delimi- tation of tracts of white matter by the process of degeneration agrees so well with the results of the embryological method as to leave no doubt that the white matter does consist of tracts which differ from each other in nature and in function. The several tracts thus indicated vary in different regions of the cord. They may be broadly described as follows. I. Descending Tracts, that is to say, tracts which undergo a descending degeneration in the sense noted above. The most important and conspicuous is a large tract (Fig. 100, cr. P.) occupying the posterior part of the lateral column, coming cr.P Cs. FIG. 100. DIAGRAM TO ILLUSTRATE THE GENERAL ARRANGEMENT OF THE SEVERAL TRACTS OF WHITE MATTER IN THE SPINAL CORD. (Sherrington.) The section is taken at the level of the fifth cervical nerve. The relations of the tracts in different regions of the cord are shewn in Fig. 104. The ascending tracts, tracts of ascending degeneration, are shaded with dots, the descending tracts, tracts of descending degeneration, are shaded with lines; the shading is in each case put on one side of the cord only the reference letters being placed on the other side. cr.P. crossed pyramidal tract, or more shortly pyramidal tract. d.P. direct pyramidal tract, shaded on the side opposite to that on which cr.P. is shaded, in order to indicate the difference of the two as to crossing. G.b. cerebellar tract, s.lr. and cr. together indicate the median posterior tract or tract of fibres of the posterior roots, cr. representing, as is explained more fully in the text, the cervical and s.lr. the sacral, lumbar and dorsal roots. asc.a.L the antero-lateral ascending tract, desc.l. the antero-lateral descending tract. The area, not shaded, marked x, is the small descending tract or rather patch mentioned in the text as observed, in certain regions of the cord, in the external posterior column rz. The small area at the tip of the posterior horn, marked L, is the posterior marginal zone or Lissauer's zone. close upon the outer margin of the posterior horn, and for the most part not reaching the surface of the cord. We shall have to return to this tract more than once, and may here simply say that it is most distinctly marked out by both the embryological and the degeneration methods, that it may be traced along the whole length of the cord from the top of the cervical region to the end of the sacral region, and that it enters the cord from CHAP, i.] THE SPINAL CORD. 873 the brain through the structures called the pyramids of the bulb, which we shall study later on. These pyramids cross over or decussate as they are about to pass into the cord, forming what is known as the decussation of the pyramids, and the tract of fibres in question shares in this decussation. Hence this tract is called the crossed pyramidal tract or more simply the pyramidal tract. A smaller, less conspicuous descending tract occupies the median portion of the anterior column (Fig. 100, d. P.) This is not only much smaller but also much more variable than the crossed pyramidal tract, is not present in the lower animals, being found in man and the monkey only and being better developed in man than in the monkey, and reaches a certain way only down the spinal cord, generally coming to an end in the thoracic region. It too comes down from the pyramid, and is a continuation of that part of the pyramid which unlike the rest does not decussate in the bulb ; thus the tract which coming down from the left side of the brain runs in the left pyramid in the bulb, passes down into the left anterior column of the cord. Hence this smaller tract is called the direct pyramidal tract. These two are the most conspicuous and important descending tracts, but names have been given to two other descending tracts. One, known as the antero-lateral descending tract, is a large tract placed in the antero-lateral column, and seen in section (Fig. 100, desc. I.) as an elongated area stretching from the py- ramidal tract towards the anterior column and reaching at times as far as the anterior fissure. The area is large, however, because the tract is very diffuse, that is to say, the fibres with descending degeneration, or fibres which degenerate below the section or injury, are very largely mixed up with fibres which do not degenerate ; in this respect this tract contrasts with the pyra- midal tract, which is to a much greater extent composed of fibres with descending degeneration, though even in it there are a considerable number of fibres which do not degenerate. Indeed this antero-lateral descending tract is so diffuse that it hardly deserves to be called a tract. The other is a small, narrow, comma-shaped tract (Fig. 100, a?), situated in the middle of the external posterior column which has been observed in the cervical and upper thoracic regions, and has been called the " descending comma tract." But the degeneration reaches a short way only, below the section or injury, and the group of fibres thus degenerating can hardly be considered as forming a tract comparable to the other tracts. The area probably represents fibres of the posterior root which take a descending course soon after their entrance into the cord. II. Ascending tracts, that is to say, tracts in which the degeneration takes place above the section or injury. A conspicuous ascending tract of a curved shape (Fig. 100, C. b.) occupies the outer dorsal part of the lateral column lying 874 THE TRACTS OF WHITE MATTER. [BOOK m. to the outside of the crossed pyramidal tract, between it and the surface of the cord. It appears to begin in the upper lumbar region, being said to be absent from the lower lumbar and sacral cord, and may be traced upwards increasing in size through the thoracic and cervical cord to the bulb. In the bulb it may be traced into the restiform body or inferior peduncle of the cere- bellum, and so to the cerebellum ; for the restiform body serves, as we shall see, in each lateral half of the brain, as the main connection of the cerebellum with the bulb and spinal cord. Hence this tract is called the cerebellar tract A second important ascending tract occupies the median portion of the posterior columns (Fig. 100, cr., s.lr.), and so far coincides with what we described above as the median posterior column, in the upper regions of the cord, that it may be called the median posterior tract ; it extends along the whole length of the spinal cord, varying at different levels in a manner which we shall presently study, and ending above in the bulb. A third ascending tract, called the ascending antero-lateral tract, or tract of Gowers, occupies (Fig. 100, asc. a. I.) the outer ventral part of the lateral column. It has somewhat the form of a comma, with the head filling up the angle left between projecting portions -of the cerebellar and pyramidal tracts, and the tail stretching away ventrally along the outer margin of the lateral column outside the antero-lateral descending column, the end of the tail often reaching to the anterior roots. It may be traced along the whole length of the cord, but is not so distinct and compact a tract as the two ascending tracts just mentioned ; the fibres with ascending degeneration, that is to say the fibres degenerating above the section or seat of injury, are very largely mixed with fibres of a different nature and origin. We may further remark that these several tracts differ from each other, in some cases markedly, as to the diameter of their constituent fibres. Thus the cerebellar tract is composed almost exclusively of remarkably coarse fibres. The median posterior tract, on the contrary, is made up of fine fibres of very equable size, while the fibres of the antero-lateral ascending tract are of a size intermediate between the other two. The pyramidal tract on the other hand is made up of fibres of almost all sizes mixed together. The tracts then which are thus marked out are, as descending tracts, the crossed and the direct pyramidal tracts, with the less distinct or important antero-lateral descending tract : and, as ascending tracts, the cerebellar tract, the median posterior tract and the less distinct antero-lateral ascending tract. If we suppose all these tracts taken away there is still left a considerable area of white matter, namely, nearly the whole of the external posterior column, the external anterior column, including the region traversed by the bundles of the anterior roots, and that part of the lateral column which lies between the antero-lateral descend- CHAP, i.] THE SPINAL CORD. 875 iiig tract and the crossed pyramidal tract on the outside and the grey matter on the inside. From this area of white matter we may put on one side at present the external posterior column because, as we shall see, this column is largely composed of the fibres of the posterior root which pass through this column, especially through the lateral part of it near the grey matter, on their way to their ultimate destination ; hence the alternative name of posterior root-zone. We may similarly leave for the present the small zone of white matter composed of very fine fibres known as the posterior marginal zone or Lissauer's zone (Fig. 100, L.), lying dorsal to the tip of the posterior horn and in the lower regions reaching to the outside of the cord ; for this too belongs to the fibres of the posterior root. Leaving these parts out of consideration we may say as regards the rest of the white matter, that the present state of our knowledge will not allow us to divide it into special tracts. All this area is largely composed of fibres which do not undergo either ascending or descending degeneration as the result of section, injury or disease. It has been suggested that these fibres either have no trophic centre at all or have double ones, one above and one below, on either of which they can in case of need lean ; so that when the fibre is divided at any level, the upper portion is still nourished from some centre above, and the lower from some centre below. At all events, whether this be the true explanation or no, the fibres in this part of the white matter cannot be differentiated into tracts by a study of their degeneration. Fibres of this kind, which we can speak of neither as ascending nor as descending, also occur in the external posterior column mingled with the fibres of the posterior root. And we may repeat the caution, that even in the several ascending and descending tracts just described, especially in those which we spoke of as less distinct or as more diffuse, many fibres are present which undergo neither ascending nor descending degeneration. § 568. It may be as well perhaps to insist here once more, that when these several tracts or the fibres running in the tracts are spoken of as ascending or descending, what is meant is that the degeneration takes place above the section or seat of injury or disease in the one case, and takes place below in the other. It has been supposed by many that the nervous impulses which these fibres severally carry, travel in the same direction as that taken by the degeneration, that the ascending tracts carry impulses from below upward, that is to say, carry impulses which arising from peripheral organs pass to various parts of the spinal cord or of the brain, that they are, in other words, channels of afferent impulses, and that conversely the descending tracts carry efferent impulses. To this view is often added as a corollary, that the tracts which do not degenerate at all carry impulses both ways, and hence cannot be considered as either afferent or efferent 876 THE NERVE ROOTS. [BOOK in. channels but simply as communicating channels. Upon this it may be remarked that impulses do not necessarily travel in the same direction as the degeneration; when a spinal nerve trunk is divided the afferent fibres as well as the efferent fibres both degenerate in a descending direction towards the periphery, though the former carry impulses in the other direction. Hence the direction of degeneration is no proof of the direction in which impulses travel ; moreover, as we have seen, degeneration does not actually travel along the fibres of the spinal cord in the same way that it does along the fibres of a nerve trunk. It may be that the descending tracts do carry impulses in a descending direction, that is, efferent impulses, and that the ascending tracts serve to carry afferent impulses ; but the proof that they do thus respectively act must be supplied from other facts than those of degeneration. Moreover, we shall have to return to these ascending and descending tracts and to study their behaviour along the length of the cord before we can use the facts concerning them as a basis for any discussion as to their functions. § 569. The connections of the nerve roots. If we regard the spinal cord, and apparently we have right to do so, as resulting from the fusion of a series of segments or metameres, each segment, represented by a pair of spinal nerves, being a ganglionic mass, that is to say a mass containing nerve-cells with which nerve fibres are connected, we should expect to find that the fibres of a spinal nerve soon after entering in, or before issuing from the spinal cord are connected with nerve-cells lying in the neighbourhood of the attachment of the nerve to the cord. We should, we say, expect to find this ; but owing to the difficulty of tracing individual nerve fibres through the tangled mass of the substance of the cord, our actual knowledge of the termination of the fibres of the posterior root, and origin of the fibres of the anterior root is at present far from complete. With regard to the anterior root, there can be no doubt that a very large proportion of the fibres in the root are continuations of the axis-cylinders of cells in the anterior horn. The fibres which can thus be traced are of large diameter and appear to be chiefly if not exclusively motor fibres for the skeletal muscles. In the frog a laborious enumeration on the one hand of the number of fibres in the anterior roots, and on the other hand of the number of cells of the anterior horn in the areas corresponding to the nerve roots has, it is true, shewn a very remarkable agreement in number between the two. We might be inclined from this to conclude that all the fibres of an anterior root start directly from cells in the anterior horn, and that all the cells in the anterior horn end in fibres of the nearest anterior root. But several considerations prevent us from trusting too much to this observation, especially in the case of the higher animals. The anterior root contains other fibres than motor fibres for the CHAP, i.] THE SPINAL CORD. 877 skeletal muscles, vaso-motor fibres for instance, secretory fibres and others ; and it is a priori unlikely that these should have origin from the same cells as the motor fibres of the skeletal muscles. Moreover, as a matter of fact some of the fibres have been traced through the anterior horn, on the one hand towards the posterior horn and on the other hand towards the lateral column; others again are found to pass through the anterior horn of their own side to the bottom of the anterior fissure where, crossing over to the other side and thus forming part of the anterior white com- missure, they appear to ascend to the anterior horn of the other side. We cannot at present make any positive statement as to the real origin and exact nature of these fibres which thus upon entering the cord pass by the cells in the anterior horn without joining them, though those which cross by the anterior white commissure are supposed to take origin in the cells of the anterior horn of the other side ; it is sufficient for our present purposes to remember that while a large number of the fibres of the anterior root, presumably those supplying the skeletal muscles, take origin in the cells of the anterior horn, shortly before they issue from the cord, others have some other origin. And similarly we have reason to think that all the cells in the anterior horn do not send out axis-cylinder processes to join the anterior roots of the same side. We may however regard a large number at all events of the cells of the anterior horn, at the level of as well as a little below and a little above the level of the exit of any particular anterior root, as constituting a sort of nucleus of origin for the larger number of the fibres, and those most probably the skeletal motor fibres, of that anterior root. The posterior root enters the cord not in several bundles laterally scattered as does the anterior root, but in a more compact mass. This mass however consists of at least two distinct bundles, which upon their entrance into the cord, take different courses. One bundle, the larger one, lying to the inner or median side of the other, consisting of relatively coarse fibres, and called the median bundle (Fig. 98, Pr'), passes obliquely into the lateral part of the external posterior column, which, as we have said, is in consequence often spoken of as the posterior root-zone. Here the fibres changing their direction run longi- tudinally for some distance upwards (some however, certainly in the upper cervical region, and probably in other regions, run a short distance downwards) but eventually either go, as we shall see, to form the median posterior tract or make their way back into the grey matter at the base of the posterior horn and thus join the vesicular cylinder, though some are said to be continued on through the grey matter into the anterior horn. The other smaller bundle placed to the outside of the former, and called the lateral bundle (Fig. 98, Pr), may be again divided into an inter- mediate bundle (Fig. 99, Pr) lying next to the median bundle, 878 THE NERVE ROOTS. [BOOK in. and into a still more lateral bundle (Fig. 99, Pr"). The former, consisting also of coarse fibres, plunges directly through the sub- stance of Rolando at the extremity of, and so into the grey matter of the horn, where the fibres changing their direction run in part at least longitudinally in the grey matter in bundles known as "the longitudinal bundles of the posterior horn" Figs. 98, 99 r. f. p. some of which appear to pass on to the anterior horn. The small most external or lateral portion of the lateral bundle, consisting of fine fibres and sometimes spoken of as the lateral bundle, on entering the cord at once ascends for some distance, and thus forms the thin layer of fine fibres, the posterior marginal zone or Lissauer's zone, indicated in Fig. 99 by ra. t.t which lies between the actual extremity of the horn and the surface of the cord, and in the upper regions of the cord (cf. Fig. 98, p') runs some way upward on the lateral margin of the horn between the grey matter and the crossed pyramidal tract. As it ascends this layer continually gives off fibres to the grey matter of the posterior horn in the cells of which they appear to end. Thus, while part of the median bundle does not join the grey matter at all but goes to form the median posterior tract, the rest of that bundle and all the other fibres of the root, sooner or later, join the grey matter either of the posterior horn or of some other part. § 570. The Special Features of the several regions of the Spinal Cord. The cord begins below in the slender filament called the filum terminale, which lying in the vertebral canal, in the midst of the mass of nerve roots called the cauda equina, rapidly enlarges at about the level of the first lumbar vertebra into the conus medullaris. This may be regarded as the beginning of the lower portion of a fusiform enlargement of the cord known as the lumbar swelling, which reaches as high as about the attachment of the roots of the twelfth or eleventh thoracic nerve at the level of the eighth thoracic vertebra, the broadest part of the swelling being about opposite the third lumbar nerve. Above the lumbar swelling, through the thoracic region the somewhat narrowed cord retains about the same diameter until it reaches the level of the first or second thoracic nerve opposite the seventh cervical vertebra where a second fusiform enlargement, the cervical swelling, broader and longer than the lumbar swelling, begins. The broadest part of the cervical swelling is about opposite to the fifth or sixth cervical nerve ; from thence the diameter of the cord becomes gradually somewhat less until it begins to expand into the bulb, but even in the highest part is greater than in the thoracic region. The sectional area of the cord increases therefore from below upwards, but not regularly, the irregularity being due to the lumbar and cervical swellings. The extremity of the filum terminale is said to consist entirely of neuroglia closely invested by the membranes, even the central CHAP, i.] THE SPINAL CORD. 879 canal being absent. A little higher up the central canal begins, and nerve-cells with nerve-fibres make their appearance in the neuroglia; thus a kind of grey matter covered by a thin super- ficial layer of white matter is established. We have already referred to the peculiar features of the lower end of the conus § 565 ; but higher up the canal becomes central and small, the posterior columns are developed, and the grey matter contains more nervous elements and relatively less neuroglia, becomes in fact ordinary grey matter. From thence onward to very near the junction with the bulb, where transitional features begin to come in, the spinal cord may be said to have the general structure previously described. The sectional area of the- white matter increases in absolute size and on the whole in a steady manner from below upwards. In other words, in a section at any level, the number of longi- tudinal fibres forming the white matter is greater than the number at a lower level, and less than the number at a higher level ; for any difference which may exist in the diameter of the individual fibres is insufficient to explain the differences in the total sectional area of the white matter. If we were to measure in man the sectional area of each of the spinal nerves as it joins the cord, and to add them together, passing along the cord from below v iv in ii i v iv in ii i x/i xi x ix viii vii vi v iv in ii i viii vii vi v iv HI ii i FIG. 101. DIAGRAM SHEWING THE UNITED SECTIONAL AREAS OF THE SPINAL NERVES, PROCEEDING FROM BELOW UPWARDS. In this as in the succeeding figures 102 — 3, — 5, — 6, — 7, all of which refer to man, the left hand side represents the bottom of the cord and the right hand the top of the cord, the numerals indicating successively the sacral, lumbar, thoracic and cervical nerves. The several figures are not drawn to the same scale. upwards the results put in the form of a curve would give us some such figure as that shewn in Fig. 101 ; the area gained by adding together the sectional areas of the nerves increases in a fairly steady manner from below upwards. The curve of the sectional area of the white matter of the cord taken from below upwards would be very similar, but if anything more regular. It must be understood however that the dimensions of the areas would not be the same in the two cases. The sectional area of the white matter at the top of the cervical region, though greater than any where lower down, is far less than the united sectional area of all the nerves below that level. The white 56 880 THE FEATURES OF DIFFERENT REGIONS. [BOOK in. matter is not formed by all the fibres from the nerves which join the spinal cord continuing to run along the cord up to the brain ; as we have seen, some at least of the fibres end in the grey matter. Nevertheless the white matter in passing up the cord appears to receive a permanent addition at the entrance of each nerve. We may infer that each nerve has a representative of itself starting from the level of its entrance and running up to some part of the brain. Whether the fibres thus representative of the nerve are continuations of the very fibres of the nerve itself, or are new fibres starting from some relay of grey matter, with which the fibres of the nerve are also connected, is another question. § 571. The grey matter in contrast to the white matter shews great variations in area along the length of the cord (Fig. 102). From the entrance of the coccygeal nerve upwards the area V IV III II I V IV III II I XII XI X IX VIII VII VI V IV III II I VIII VII VI V IV III II I FlG. 102. DlAGBAM SHEWING THE VARIATIONS IN THE SECTIONAL AREA OF THE GREY MATTER OF THE SPINAL CORD, ALONG ITS LENGTH. increases very rapidly, reaching a maximum at about the level of the 5th lumbar nerve. It then rapidly decreases to about the level of the llth thoracic nerve, maintains about the same dimensions all through the thoracic region, and begins to increase again at about the level of the 2nd thoracic nerve. Its second maximum is reached at about the level of the 5th or 6th cervical nerve, after which the area again becomes smaller, remaining however at the upper cervical region much larger than in the thoracic region. The meaning of these variations becomes clear when we turn V IV III H I V IV III II I XII XI X IX VIII VII VI V IV III