Columbia ©nitif m'tp mtl)e(£ttpofigfmg0rk CoUcgc of ^fipsiicians anb ^urgcong l.ii)rarp Digitized by the Internet Archive in 2010 with funding from Open Knowledge Commons (for the Medical Heritage Library project) http://www.archive.org/details/humananatomyincl02pier HUMAN ANATOMY INCLUDING STRUCTURE AND DEVELOPMENT AND PRACTICAL CONSIDERATIONS BY THOMAS DWIGHT, M.D., LL.D. J. PLAYFAIR McMURRICH, PH.D. PARKMAN PROFESSOR OF ANATOMY IN HARVARD PROFESSOR OF ANATOMY IN THE UNIVERSITY OF UNIVERSITY MICHIGAN CARL A. HAMANN, M.D. GEORGE A. PIERSOL, M.D., SC.D. PROFESSOR OF ANATOMY IN WESTERN RESERVE PROFESSOR OF ANATOMY IN THE UNIVERSITY OP UNIVERSITY PENNSYLVANIA AND J. WILLIAM WHITE, M.D., PH.D., LL.D. JOHN RHEA BARTON PROFESSOR OF SURGERY IN THE UNIVERSITY OF PENNSYLVANIA WITH SEVENTEEN HUNDRED AND THIRTY-FOUR ILLUSTRATIONS, OF WHICH FIFTEEN HUNDRED AND TWENTY-TWO ARE ORIGINAL AND LARGELY FROM DISSECTIONS BY JOHN C. HEISLER, M.D. PROFESSOR OF ANATOMY IN THE MEDICO-CHIRURGICAL COLLEGE EDITED BY GEORGE A. PIERSOL VOL. n. THIRD EDITION PHILADELPHIA & LONDON J. B. LIPPINCOTT COMPANY Coryright. 1906. by J. B. Lippincott Company. Copyright. 1907, by J. B. Lippincott Company. Copyright, 190S, by J. B. Lippincott Company. Copyright, 191 1, by J. B. Lippincott Company. Entered at Stationers' Hall, London, England. All Rights Reserved. IieeTROTVPEO *NOPRINTEOaV<.B. LIPPINCOTT COMPANY, PHILADELPHIA, M .%*. CONTENTS, \OL. II. THE NERVOUS SYSTEM. General Considerations 996 The Nervous Tissues 997 The Nerve-Cells 997 The Nerve-Fibres 1000 Neuroglia 1003 The Nerve-Trunks 1006 The Ganglia 1007 Development of the Nervous Tissues . . . 1009 Nerve-Terminations 1014 Motor Handings 1014 Sensory Endings 1015 The Central Nervous System. The Spinal Cord Membranes Cord-Segments Form of the Spinal Cord Columns of the Cord Gray Matter Central Canal Microscopical Structure White Matter Fibre Tracts Blood-Vessels of Spinal Cord Development of Spinal Cord Practical Considerations : Spinal Cord. . Malformations Injuries Localization of Lesions The Brain General Description General Development Derivatives from the Rhombencephalon The Medulla Oblongata Internal Structure The Pons Varolii Internal Structure The Cerebellum Lobes and Fissures Architecture Internal Nuclei Cerebellar Cortex Cerebellar Peduncles The Fourth Ventricle Development of the Hind-Brain Derivatives The Medulla The Pons The Cerebellum The Mesencephalon The Corpora Quadrigemina The Cerebral Peduncles The Sylvian Aqueduct Internal Structure of the Mid-Brain The Tegmentum The Crusta The Median Fillet The Posterior Longitudinal Fas- ciculus 102 1 1022 1024 1026 1027 1028 1030 1030 1036 1039 1047 1049 105 1 105 1 1052 1053 1055 1056 1058 1063 1063 1068 1077 1078 1082 1084 1088 1088 1090 1093 1096 IIOO IIOI 1 103 1 103 1 105 1 106 1 107 1 108 III2 III2 III5 III6 The Mesencephalon — Continued Development of Mid-Brain 1117 The Fore-Brain 1 119 The Diencephalon 11 19 The Thalamus 1 1 19 Structure 11 20 Connections 1121 The Epithalamus 11 23 The Trigonum Habenulae 11 23 The Pineal Body 11 24 The Posterior Commissure 1125 The Metathalamus 11 26 The Hypothalamus 11 27 The Subthalamic Region 11 27 The Corpora Mammillaria 1128 The Pituitary Body 1129 The Third Ventricle 1131 The Telencephalon 1132 The Cerebral Hemispheres 1133 Cerebral Lobes and Interlobar Fissures 1 135 Lobes of the Hemispheres 1139 Frontal Lobe 1 139 Parietal Lobe 1143 Occipital Lobe 1 145 Temporal Lobe 1 147 Insula 1 149 Limbic Lobe 1 150 The Rhinencephalon 1151 The 01factor>- Lobe 1 151 Architecture of the Hemispheres 1 155 The Corpus Callosum : . . . 1155 The F"ornix 1 158 The Septum Lucidum 1159 The Lateral Ventricles 1160 Internal Nuclei of the Hemisphere 1169 The Caudate Nucleus 1 169 The Lenticular Nucleus 11 69 The Claustrum 11 72 The Amygdaloid Nucleus 1172 The Internal Capsule 1 173 Structure of the Cerebral Cortex 1175 The Nerve-Cells of Cortex 1 176 The Nerve-Fibres of Cortex. ....... 11 79 Variations in Cerebral Cortex 1180 White Centre of the Hemisphere 1182 The Association Fibres 1182 The Commissural F"ibres 1 184 The Projection Fibres 1187 Development of the Derivati\es of Fore- Brain 1 189 The Pallium 1189 The Sulci and Gyri 1190 Histogenesis of Cerebral Cortex. .. . 1192 The Rhinencephalon 1195 The Corpus Striatum 1 193 The Diencephalon 1193 The Cerebral Commissures 1194 Measurements of the Brain 1195 VI CONTKNTS. PACE The Membranes of the Brain 1197 TIk- Dura INIatt-r 1 19.S The I'ia Mater 1202 The ArachnoitI 1203 The Pacchionian Bodies 1205 The BloocIA'essels of the l.rain 1206 Practical Lonsicleratit)ns : The Brain and " Its Membranes 1207 Conjjenital Krrors of Development. . 1207 The Menini^es 120S Cerebral Hemorrhajje 1209 Cerebral Localization 12 10 Cranio-Cerebrai Topofjraphy 12 14 The Pekiphekai- Nervous Svstem. The Cranial Nerves 1220 The Olfactory Nerve 1 220 The Optic Nerve 1223 The Oculomotor Nerve 1225 The Trochlear Nerve 1228 The Trigeminal Nerve 1230 The Gasserian Ganglion 1232 The Ophthalmic Nerve and Branches 1233 The Ciliar\- Ganglion 1236 The Maxillary Nerve and Branches 1237 The Spheno-Palatine Gang- lion 1240 The Mandibular Nerve and Branches 1242 The Otic Ganglion 1246 The Submaxillary Ganglion 1247 Practical Considerations : The Tri- geminal Nerve 1 248 The Abducent Nerve 1249 The Facial Nerve 1250 Practical Considerations 1254 The Auditory Nerve 1256 The Glosso-Pharj-ngeal Nerve 1260 The Vagus or Pneumogastric NerVe 1265 Practical Considerations 1272 The Spinal Accessory Nerve 1274 Practical Considerations 1275 The Hypoglossal Nerve 1275 Practical Considerations 1277 The Spinal Nerves 1278 The Posterior Primary Divisions 1279 The Cervical Nerves 1281 The Thoracic Nerves 1282 The Lumbar Nerves 1282 The Coccygeal Nerve 1284 The Anterior Primary Divisions 1284 The Cervical Nerves 1285 The Cervical Plexus and Branches 1286 The Phrenic Nerve 1290 Practical Considerations 1292 The Brachial Plexus and Branches 1292 The External Anterior Thoracic Nerve 1 297 The Musculo-Cutaneous Ner\'e 1 298 The Median Nerve 1298 PACK The Brachial Plexus and Branches— Co»/i;iued Practical Considerations 1301 The Internal Anterior Thoracic Nerve 1303 The Lesser Internal Cutaneous Nerve 1 303 The Internal Cutaneous Nerve 1303 The Ulnar Nerve 1303 Practical Considerations 1306 The Subscapular Nerves 1306 The Circumlkx Nerves 1307 Practical Considerations 1308 The Musculo-Spiral Nerve 1308 Practical Considerations 1314 The Thoracic Nerves 1314 Practical Considerations 1318 The Lumbar I'lexus and Branches 1319 The llio-Hypogastric Nerve 1320 The llio-lnguinal Nerve 1321 The Genito-Crural Nerve 1322 The External Cutaneous Nerve 1324 The Obturator Nerve 1324 The Accessory Obturator Nerve. .. . 1324 The Anterior Crural Nerve 1327 Practical Considerations : Lumbar Plexus 1330 The Sacral Plexus and Branches 1331 The Great Sciatic Nerve 1335 The External Popliteal Nerve 1336 The Anterior Tibial Nerve 1336 The Musculo-Cutaneous 1338 The Internal Popliteal Nerve 1339 Tiie Posterior Tibial Nerve 1342 The Pudendal Plexus and Branches 1345 The Small Sciatic Nerve 1348 The Pudic Nerve 1349 The Coccygeal Plexus 1352 Practical Considerations : Sacral Plexus 1352 The Svmpathetic Nerves 1353 General Constitution and Arrange- ment 1355 The Gangliated Cord 1356 Rami Communicantes 1356 Cervico-Cephalic Portion of Gangliated Cord 1358 The Superior Cervical Gan.glion 1359 The Middle Cervical Ganglion 1362 The Inferior Cervical Ganglion 1362 Thoracic Portion of Gangliated Cord. . . . 1364 The Splanchnic Nerves 1364 Lumbar Portion of Gangliated Cord. .. . 1366 Sacral Portion of Gangliated Cord 1367 The Plexuses of the Sympathetic Nerves 1367 The Cardiac Plexus 1367 The Solar Plexus 1368 Subsidiary Plexuses 1369 The Hypogastric Plexus 1374 Subsidiary Plexuses 1374 Practical Considerations : The Sympa- thetic Nerves 1375 Development of the Peripheral Nerves. . 1375 THE ORGANS OF SENSE. The Skin. General Description 1381 Structure 1382 The Hairs 1389 Structure 1391 The Nails i394 The Cutaneous Glands i397 The Sebaceous Glands 1397 The Sweat Glands 1398 Development of the Skin and its Append- ages i4otJ CONTENTS. vi; i'Af;r'; TllK NOSK. Tlie Outer Nose 1404 Cartilaj^es of the Nose 1404 Practical Considerations : The External Nose 1407 The Nasal Foss:e 1409 The Vestibule 1409 The Septum 1410 The Lateral Wall 1410 The Nasal Mucous Membrane 1413 The Olfactory Region 1413 The Respiratory Region 1415 Jacobson's Organ 1417 Practical Considerations : The Nasal Cavities 1417 The Accessory Air- Spaces 1421 The Maxillary Sinus 1422 The Frontal Sinus 1423 The Ethmoidal Air-Ceils 1424 The Sphenoidal Sinus 1425 Practical Considerations : The Accessory Air-Spaces 1426 Development of the Nose 1429 The Organ of Taste The Taste-Buds 1433 Structure 1434 Development 1436 The Eye. The Orbit and its Fasciae 1436 Practical Considerations 1438 The Eyelids and Conjunctiva 1441 Practical Considerations 1446 The Eyeball 1447 Practical Considerations 1448 The Fibrous Tunic 1449 The Sclera 1449 The Cornea 1450 Practical Considerations 1453 The Vascular Tunic 1454 The Choroid 1455 The Ciliary Body 1457 Practical Considerations 1459 The Iris 1459 Practical Considerations 1461 The Nervous Tunic 1462 The Nervous Tunic — i'oiitinucd The Retina Practical Considerations The Optic Nerve I'ractical Considerations The Crystalline Lens Practical Considerations The Vilrecjus BiTdy Practical Considerations The Suspensory Aj^ixuatus of the Lens. The Acjueous Humor and its Chamber. Practical Considerations The Lachrymal Apparatus The Lachrymal Gland The Lachrymal Passages Practical Considerations Development of the Eye The Ear. The External Ear The Auricle The External Auditory Canal Practical Considerations The Middle Ear The Tympanic Cavity The Alembrana Tympani .... The Auditory Ossicles The Mucous Membrane The Eustachian Tube The Mastoid Cells Pract. Consid.: The Middle Ear The Tympanic Cavity The Tympanic Membrane The Eustachian Tube The Mastoid Process and Cells. . . . The Internal Ear The Osseous Labyrinth The Vestibule The Semicircular Canals The Cochlea , The Membranous Labyrinth , The Utricle The Saccule The Semicircular Canals The Cochlear Duct , The Nerves of the Cochlea Development of the Ear 462 468 469 470 471 473 473 474 475 476 476 477 477 478 479 480 484 487 490 492 492 494 496 500 501 504 504 504 505 507 508 510 511 511 512 513 514 514 515 515 517 521 523 THE GASTRO-PULMONARY SYSTEM. General Considerations 1527 Mucous Membranes 1528 Structure 1528 Glands. . : 1531 Types of Glands 1531 Simple Tubular Glands 1532 Compound Tubular Glands .... 1532 Tubo- Alveolar Glands 1532 Serous Glands 1534 Mucous Glands 1534 Simple Alveolar 1535 Compound Alveolar Glands. .. . 1535 Development of Glands 1537 The Alimentary Canal. The Mouth 1538 The Lips, Cheeks and Vestibule 1538 The Teeth 1542 Description of Individual Forms. .. . 1543 Structure of the Teeth 1548 The Enamel 1548 The Dentine 1550 The Teeth — Cotitinued The Cementum 1552 The Alveolar Periosteum 1553 Implantation and Relations of the Teeth 1554 Development of the Teeth 1556 First and Second Dentition 1564 The Gums 1567 The Palate 1567 The Hard Palate 1567 The Soft Palate 1568 The Tongue 1573 General Description 1573 The Glands of the Tongue 1575 The Muscles of the Tongue 1577 The Sublingual Space 1581 The Salivary Glands 15S2 The Parotid Gland 1582 The Submaxillary Gland 1583 The Sublingual Gland 1585 Structure of the Salivary Glands. . . . 1585 Development of the Oral Glands. . . . 1589 vm CONTENTS. I'AOK Practical Considerations : The Month . . . 1 589 Malformations: Harelip and Cleft Palate 1589 The Lips 1590 The Ciums 1590 The Teeth 1591 The Koof of the Mouth 1592 Tlie Floor of the Mouth 1593 The Cheeks 1 594 The Tongue 1594 The Pharynx 1596 The Naso- Pharynx 1598 The Oro-Pharynx 1598 The Larynj^o-Pharynx 1598 The Lymphoid Structures 1599 The l'"aucial Tonsils i6cxj The Pharyngeal Tonsil 1601 Relations of the Pharynx 1601 Development and Growth of Pharynx 1603 Muscles of the I^harynx 1604 Practical Considerations : The Pharynx. . 1606 The Qisophaj;us • 1 609 General Description 1609 Course and Relations 1609 Structure 161 1 Practical Considerations : CEsophagus . . 1613 Congenital Malformations 1613 Foreign Bodies 1613 Strictures 1614 Carcinoma 1614 Extrinsic Disease 1614 Diverticula 1614 The Abdominal Cavity 1615 The Stomach 1617 General Description 1617 Peritoneal Relations 1619 Position and Relations 1619 Structure 1621 Growth 1629 Variations 1629 Practical Considerations : The Stomach 1629 Congenital Malformations 1629 Injuries of the Stomach 1630 Ulcers and Cancer 1631 Dilatation and Displacement 1631 Operations on the Stomach 1632 The Small Intestine 1633 General Description 1633 Structure 1634 The Duodenum 1644 Duodeno-Jejunal Fossa; 1647 Interior of the Duodenum 1648 The Jejuno-Ileum 1649 The Mesentery and Topography .... 1650 Meckel's Diverticulum 1652 Practical Considerations: The Small In- testine 1652 The Peritoneal Coat 1652 The Muscular Coat 1653 The Mucous and Submucous Coats 1653 Ulcers of the Duodenum 1653 Infection 1654 Typhoid Ulcers 1654 Contusion and Rupture 1654 Obstruction 1655 Operations 1656 The Large Intestine 1657 General Description 1657 Structure 1657 The Ccecum 1660 The Vermiform Appendix 1664 Peritoneal Relations 1665 Pericaical Foss:e 1666 The Large Inlest'inc^ Co>//inut'(/ Retro-Colic Fossu; 1667 The Colon 1668 General Description i66• Carrier, Holmes and others it seems probable that they are normal constituents of the cell and are directly related to functional activity, undergoing increase under unusual stimulus. The intracellular canals described by Holmgren as existing in nerve-cells, in connection with a reticulum {h-ophospofigiuin) that appears after certain treatment, have been variously interpreted. By not a few they are regarded as artefacts, or at least dependent upon the intra- cellular fibrilke for their exhibition. Pewsner-Neufeld, however, believes them to be lymph- c:lefts within the cytoplasm that directly communicate with lymph-spaces which surround the nerve-cell and tinis provide a means for the rapid removal of waste products from the neurone. Fig. S39. Every neurone possesses at least one process, which is then an axone, although usually provided with both dendrites and axone. Very rarely more than a single axone is present. Depend- ing upon the number of their processes, nerve-cells are described as unipolar, bipolar., or multipolar. The unipolar condition is often secondary, since two processes may be so blended for part of their course that they form a single process. Conspicuous examples of such relation are seen in the spherical nerve-cells composing the spinal and other ganglia connected with the sensory nerves. Primarily such neurones possess an axone and a dendrite that arise from opposite ends of what is for a time a spindle-shaped bipolar cell. During development, however, the unilateral growth of the cell-body towards the surface of the ganglion brings about the gradual approximation of the two processes until they fuse in the single extension into which the spherical or flask-like cell is prolonged. This process sooner or later undergoes a Y- or T- like division, one process, usually identified as the dendrite, passing to the periphery to end in the free terminal arborization, whilst the other, the axone, passes centrally to end in an arborization around the neurones lying within the cerebro-spinal axis. Examples of bipolar neurones, in which the dendrite and axone pass from opposite sides of the spherical cell-body, are found in the retina and the ganglia Diagram showing transformation of young bipolar sensory neurone into one of unipolar type. HUMAN ANATOMW Fk;. S40. connected with the acoustic nerve. An interestinj; mochtication of l)i])()lar neurones is presented hv the olfactory cells, whose dendrites are represented by the extremely short processes embedded within the nasal mucous membrane, whilst the axones are prolonfj;ed as the fil)res of the olfactory nerves into the cranial cavity to end in telodendria within the glomeruli (jf the olfactory bulb. The cell-bodies of the multipolar neurones, which possess one axone and several dendrites, vary in form (Fig. 841). Some, as those within the sym- pathetic ganglia, are approximately spherical and of moderate size, with short delicate dendrites ; many are of large size and irregularly stellate form, the dendrites passing out in all directions, as seen in the conspicuous motor neurones within the gray matter of the spinal cord ; others possess a regular and characteristic form, as the Hask-shaped cells of Purkinje within the cerebellum, or the pyramidal cells of the cerebral cortex. Certain multipolar neurones within the cerebral cortex, and especially those constituting the chief components of the granule layer, of the cerebellum, are distinguished by the small size of their cell-bodies and the peculiar ramitications and claw-like telodendria of their dendrites (Fig. 945 j. Within the cerebellar cortex are likewise found examples of Bipolar neurones ; a, from olfactory mucous membrane — dendrite is above; f>, from retina. ^Modified fiom CajaL) Fig. S41. Multipolar nerve-cells of various forms; .1, from spinal cord ; B. from cerebral cortex; C, from cerebellar cortex (Purkinje cell) ; ii, axone; r, ini|)lantation cone. the multipolar neurones of Golgi's type II, whose axones almost immediately undergo elaborate branching within the gray matter to which they are confined. The Nerve-Fibres. — From the foregoing considerations it is evident that the nerve-fibres are not independent elements, but that all are the processes of neurones — either the axones of those that are prolonged into fibres (type I), or the dendrites of those situated within the spinal and other sensory peripheral ganglia. Although neurones exist which are not continued as nerve-fibres, the latter are alwa^•s connected rill-: Ni:R\()i s rissuKS. lOOI Im(;. S42. Axis-cvliiulers Axolcniina Medullary sheath Node of Kaiix let' Xeutileinma Medullated iierve-fibres, as seen in longi- tudinal sections of spinal nerve. X 500. with iK-iirones. Reconniziiit;, tin ix-loic, that thi- ncr\'c-fibres arc only processt-s ot lU'Uioiics, their separate descrijition is justified only as a matter of convenience. The fmulaniental |)art of every nerve-tihre is the central cord, coinnKjnly known as the axis-cylinder, which is composed of threads of great delicacy, the ax/s- fibrillce, prolonged from the nerve-cell and embcdtled within a semifluid interfihrillar substance, the neuroplasm, the entire cord so con- stitiiteil being enclosed by a delicate structiweless sheath, the axolcniina. The existence of the axolcmma as a distinct sheath, lu)vvever, is ciues- tionable, the appearance of such investment not improhal)lv being due to a local condensation of the framework of the medullary coat immediately around the axis-cylinder. In the case of the typical fibres, such as form the chief constituents of j^the peripheral nerves distributed to various parts of the body, the axis- cylinder is surrounded by a relatively thick coat, known as the medulla ly sheath, outside of which lies a thin structureless envelope, the neurilemma or sheath of Schwann, that invests the entire ner\e-hbre. In the case of fibres proceeding from neurones composing the sensory ganglia, the neurilemma is continuous with the nucleated sheath enclosing the individual ganglion-cells. The medullary sheath consists of two parts, a delicate Yi^lxcuhw framc7cork and a fatty substance, the myelin, that fills the meshes of the supporting reticulum. The latter, arranged for the most part as anastomosing membranous lamelhe, that in transverse sections of the nerve-fibre appear as faint concentric lines, resists pancreatic digestion and fat-dissolving reagents, and was regarded by Evvald and Kiihne as possessing properties similar to the keratin of horny substances and, hence, was named by them neurokeratin. The blackening after treatment with osmic acid and other reactions exhibited by myelin indicate its fatty nature, and it is probable that this substance exists during life in the form of a fine emulsion supported by the framework. When fresh, myelin appears highly refracting and homogeneous, and confers upon the medullated nerve-fibres their characteristic whitish color. It is, however, prone to post-mortem changes, so that after death it loses its former uniformity and presents irregular contractions and collections, or at the broken end of the fibre extrudes in irregular globules, due probably to fusion of the normal individual minute droplets into larger masses. The medullary sheath is not uniformly continuous, but almost completely inter- rupted at regular, although in different fibres variable, intervals marked by annular constrictions. These constrictions, the nodes of Ranvier, correspond to narrow zones at which the medullary sheath is practically wanting and the neurilemma dips in and, some- what thickened, lies in close relation with the axis-cylinder. According to Hardesty ' the medullary sheath does not suffer complete suppression at the nodes, but is represented by part of its reduced framework which trans- verses the constriction, a conclusion which we can confirm. The nodes occur at regular intervals along the fibre, which they thus divide into a series of internodal segments. In general, the latter are longer in large fibres, where they have a length of about i mm., and shorter in those of small diameter, in which they may measure . i mm. or less in length. The axis-cylinder passes uninter- ruptedly across the nodes, although it often presents a slight fusiform enlargement .Axis-cylinder Neurilemma Medullary sheatli Medullated nerve-fibres in transverse section. X 550. 'Amer. Journal of Anatomy, vol. iv., 1905. I002 HUMAN ANATOMY. opposite each constriction (Ranvier). Tlie neurilemma also suffers no break at the nodes, but is continucjus from one segment to the other. In atklition to the partial inlerru|jtions at the nodes, the medullary sheath after treatment with osmic acid freciuently apjK-ars broken by clear narrow clefts that e.vtend obliquely from the neuri- lemma to the a.xolemma and thus subdivide each internodal segment into a number of smaller tracts, known as the Schmidt- 1 .antcrmann segments (Fig. 844). The oblique clefts do not all extend in the same direction, even within the same inter- nodal segment, since they are usually directed from without inward and towards the nodal constrictions and, therefore, have an opposed tlisposition at the ends of the same as well as of the adjoining seg- ments. The significance of this subdivision is un- certain ; many regarding the details as artefacts. According to Capparelli ', however, the apparent clefts are in reality unstained membraneous septa that pass obliquely from the a.xolemma to the inner surface of the neurilemma and serve to hokl the a.xis-cylinder in place and to enclose the myelin. The studies of Hatai - on the arrangement of the neurf)keratin seem to support these conclusions. Within each internodal segment, beneath the sheath of Schwann, lies a single (sometimes more than one) small neurilemma-cell which consists of an elongated oval nucleus surrounded by a meagre amount of cytoplasm. These cells represent the remains of the mesoblastic elements {sheath-cells) that during the growth of the nerve-fibre were acti\e in i)roviding its envelope (page iciij. Fig. 845. Node of Raiivier Medullated nerve-fibres after treatment with osmic acid; A, fibre showing reticu- lum within medullar)- coat ; li, one showing same coat divided into segments. ,< 500. Medullated nerve-fibres becoming nonmedullated on a|>i>roachiiig their termination. •; 235. Depending ujjon the presence or absence of the medullary sheath throughout the greater part of their course, nerve-fibres are distinguished as medullated or non- ' Archiv f. mikros. Anat. u. Kntuick., Bd. 66, 1905. • ^Journal of Comparative Neurology, vol. xiii., 1903. THE NERVOIS TISSUES. lOo: medullated. The medullated fibres con.siiiutc tlic orcuL iiKijoriiy of those inakiiii; up the peripheral nerves and the tracts of the cerehro-spinal axis ; the component fibres of the latter, however, while niedullated are without the neurilennna. The nonmedullated fibres, on the other hand, are chiefly prolongations (axones) from the gaui^lion cells of the sympathetic system, althouj^h in the case of the olfactory nerves the fibres are also without a myelin-coat. The dis- tinction between these two classes of fibres is relative rather than Fu;. .S46. absolute, since every medullated nerve-fibre becomes nonmed- ullated before reaching its termination, central or ])erij)heral. ' ' ,4 Medullary nerve-tibres vary greatly in thickness, tiie smallest hav- ing a diameter of only .001 mm., whilst the largest may measure as much as .020 mm. According to their diameter, as determined by Ivolliker, the medullated libres may be grouped as fine (.002-.004 nun.), medium (.005-. 009 nun.), and coarse (.010-. 020 mm.), hi general, the thicker fibres are the longer and are the processes of large nerve-cells ; conversely, the finer have shorter courses and belong to small cells. Although subject to many exceptions, the motor fibres are usually the thicker and the .sensory the smaller. Since there are many more nerve-fibres than nerve-cells, it is evi- dent that the former must underg5 division along their course. Such doubling always occurs at a point corresponding to a node of Ranvier, never within the internodal .segment, the sheaths being continued over the two resulting fibres. On approaching their peripheral termination tiie branching becomes more frequent and the medullary sheath thinner until it ends, after which the axis-cylinder continues invested with only the attenuated neurilemma. The latter, now reduced to an extremely delicate covering beset with occasional nuclei, sooner or later disappears, the naked axis-cylinder alone being prolonged to end finally in the varicose threads of the telodendriun. The nonmedullated nerve-fibres proper, also tenwed pa /e Jid res ox fibres of Reniak, include those that are without the myelin sheath throughout their course. They are chiefl\- the axones of sympathetic neurones. Devoid of medullary sheath, these fibres, often .002 mm. or less in diameter, consist of only the axis-cylinder and the neurilemma, the latter being thinner antl more delicate than on the medullated fibres. Like the latter, the pale fibres end in telodendria composed of naked axis-cylinders, bearing irregular varicosities. ^^^ Nonmedullated nerve- fjbres in longitudinal section ot splenic nerve. ,■: 310. Fig. S47 Neuroglia.— The neurones (nerve-cells and fibres) within the cerebro-spinal axis are e\erywhere held together by a special supporting- tissue known as neuro^-lia. The latter is primarily derived from the invagi- nated ectoblast lining the neural tube, certain elements, the spongioblasts, being de\'Oted to the production of the neuroglia, while others, the neuroblasts, give rise to the neurones. At first the supporting tissue is represented by greatlv elongated, radially disposed fibre-cells that often extenci the entire thickness of the wall of the neural canal. Later, the neurogliar elements become differentiated into (a) those bordering the lumen of the canal, which are partly retained as the cpendymal cells, and {b') those which have early migrated to more peripheral locations and given rise to stellate cells that are converted into spider-like elements, the astrocytes. Seen in chrome-silver preparations (Fig. 847) these appear as irregular triangular or quadrilateral cells from whose angles numerous delicate fibrillee extend between the surrounding nervous elements. According to Rubaschkin,^ the astro- cytes are transformations from larger branched gliogenetic cells, by the conversion of whose robust protoplasmic processes the delicate fibrillcc that later form the chief Vouns neuroglia cells ; astrocytes of child. X 300. from brain 'Archivf. mikros. Anat. u. Entwick., Bd. 64, 1904. icx)4 HUMAN ANATOMY. constituents of the neuroglia arise. So long as neuroglia is being produced, as in the nervous axis of young animals, the large gliogenetic cells are present and directly concerned in the production of additional fibrilht, their cytoplasm becoming pro- gressi\ely less granular and reduced through the various transition phases until in the final condition, as the small ^s^/ia cc/Zs, little more than the nucleus remains. During these changes very many tibriliie lose their connection with the cells and, in conjunction with the glia threads still attached to the astrocytes, form an elaborate interlacement in which the neuroglia cells, now reduced and for the most part devoid of jM-ocesses, lie scattered at uncertain inter\als. In all parts of the central nervous system the mature neuroglia consists of essentially the same tissue, the differences presented in certain localities depending largely upon variations in its compactness. Everywhere the chief part of the sup- porting tissue consists of the intricate felt-work of fibrilhe, glia-fibrcs, as they are called, which are usually free but to some extent connected with the spidcr-cclls or astrocytes. Where, howe\er, the neuroglia borders the neural tube ( the \entricles of the brain and the central canal of the spinal cord ) as the cpoidywal layer, its arrangement exhibits peculiarities tliat call for later special mention. In the immediate vicinity of the neurones the felt-work of the fibrillie is unusually close, so that the cell-bodies and the roots of the processes are surrounded by a protectinjj sheath, the glia-capsule. This diminishes alonjj the dendrites, and after these begin to branch tlie neuroglia no longer forms a complete special investment. The medullated nerve-libres within the brain and spinal cord are also provided with delicate neuroi^Iiar sheaths which replace the neurilemma which on these fibres is wanting. These sheaths are prolonged for some distance on the fibres of the roots of the spinal nerves. The fibres of the optic nerve and of the olfactory tract are accompanied through- out their length by neurogliar sheaths, those of the remaining cranial nerves losing these envelopes shortly after leaving the brain (Rul)aschkin), Beneath the i)ia mater the neuroglia is especially dense and forms the e.xternal subpial layer that every- where invests the nervous mass, following all the inequali- ties of its surface. In this manner the pia mater is excluded and, except where its connective-tissue strands accompany the blood-vessels that enter the nervous mass, takes no part in tlie make-up of the supporting stroma. The subpial layer consists of a dense felt-work of glia-fibres, disposed in various planes, which are partly free and partly the processes of spider cells. Internally the layer fades into the adjoining diffuse neuroglia without demarcation. At the periphery- the fibres often exhiiiit a radial disposi- tion, their outer ends usually being somewhat expanded. Within the white matter the neuroglia, both in its distri- bution and density, is fairly uniform, although special tracts often .sejxirate the larger bundles of nerve-fibres. Its arrangement within the gray matter jiresents less uniformity, since more or less marked condensations occur where the nerve-cells are collected into nuclei, as conspicuously seen in the inferior olive. Where the neuroglia borders the neural tube (especially the central canal of the sj^inal cord) it constitutes the ependymal layer, the peculiari- ties of which call for special mention. The imme- diate lining of the tube consists of a single layer of pyramidal epithelial elements, the ependymal cells, whose free siu-faces or bases look towards the lumen, and the apices towards the surrounding nervous tissue. At least during the earlier years in man, and throughout life in many lower mammals, the free surface of each cell is beset with a number of hair-like processes that in their relations with the cytoplasin correspond to ordinary cilia. The pointed distal end of the ependymal cell is prolonged into a conical process that is directly continued into usually a single neurogliar fibre which, after a course of uncertain length becomes Ependymal cells and adjacent neuro- glia surrounding central canal of spinal cord of cat. X 75. (Kubaschkin.) THE NERVOUS TISSUES. 1005 lost in the surroundinsf complex of j^lia-hbrcs. In young tissue the apical processes often exhibit e\ iclences of l)reakinj4 up into a number of tine librilLt. Where the processes enter robust tracts of nemoglia, as in the posterior longitudinal septum of the spinal cord, they are of unusual length. In addition to the radially directed fibres connected with the ependymal cells, the tibre-comi^lex of the ependymal zone includes many fibrillae that are circularly and longitudinally disposed. Scattened glia cells, some stellate but mostly small, are also present and represent the elements from which the neuroglia-fibrilhe ha\'e been derived. hi tlie preceding account of the elements cunipusiiij; the nervous tissues the neurones have been regarded as the niorphoiogical units, each retaining its individual anatomical indepen- dence, although functionally closely related with other similar units. This conception, com- monly referred to as the Neurone Doctrine and strikingly formulated by VV'akleyer in 1S91, stands in contrast to the prior views by which actual continuity was attributed t'■'.' • -'' ,- ~^^'_iy!/{' y,. '""■"}• '''"' — '"^ Perineurium .Funiculus of nerve-fibres Transverse section of small nerve-trunk composed of loosely united funiculi. V 20. The latter, lined by flattened connective-tissue plates, are in relation with the clefts between the nerve-fibres, on the one hand, and with the lymphatics within the inter- funicular tissue on the other. Where, as usual, the nerve is composed of se\eral funiculi, these are loosely bound together and the entire trunk so formed is invested by a general fibro-elastic envelope, the cpmetiriiim, in which course the blood-vessels and lymphatics. These envelopes of the nerve-trunk are continued over its branches, even onto its smallest subdivisions. The last representative of these coverings is seen on the individual fibres as the sheath of Henle, that surrounds the fibre and consists of flattened cells and delicate strands of connective tissue outside the neurilemma. In cross-sections of the nerve-trunk (Fig. 850), the transversely cut individual meduUated nerve-fibres appear as small circles, sharply defined by a fine outline (the neurilemma), each enclosing a deeply stained dot (the axis-cylinder in section). The interval between the latter and the neurilemma, corresponding to the space occupied by the myelin, usually appears clear and unstained with the exception of delicate and uncertain suggestions of membranous septa. In contrast with its unstained appearance in sections tinged with carmine, after the action of osmic acid or special hematoxylin staining (Weigert) the medullary substance exhibits a dark color and the axis-cylinder appears surrounded by a deeply tinted ring. The neuri- THE NERVOUS TISSUES. 1007 lemma nuclei are occasionally seen as deeply stained crescentic figures that partially embrace the nerve-fibre, lying beneath the neurilemma within depressions in the medullary substance. Fig. 850. -0. (ST'.- • Perineurium — ■ ' - .-- ■■■■.-.-> 'iJ'iRy « Endoneurium >;i:;>:-U:*';^:^^^^ Nerve-fibre Epineurium .lg^ Blood-vessel Transverse section of funiculus composed of nerve-fibres held together by endoneurium and surrounded by perineurium. X 175. Viewed in cross-section, the nonmeduUated fibres appear as small irregularly round figures arranged in groups that correspond to bundles (Fig. 851). When numerous, the latter are aggregated Fig. 85 1. Epineurium 'r--- ;>\\. \ ):^l into secondary bundles between which extend delicate connective- tissue septa, continuous with the general envelope investing the nerve- trunk. The medullary substance being wanting, the pale fibres are of small size and often possess a diameter of less than .001 mm. The Ganglia. — The cell- bodies of the neurones that consti- tute the sensory pathways within the peripheral nerves and of the neu- rones of the sympathetic system are collected at various points into aggregations known as ganglia. Familiar examples of the latter are the spinal ganglia on the posterior roots of the spinal nerves, certain cranial ganglia (as the Gasserian connected with the fifth nerve, the acoustic with the eighth, and those on the trunks of the seventh, ninth and tenth cranial nerves), and the sympathetic ganglia along the gangliated cords and within various plexuses of the sympathetic. A longitudinal section of a spinal ganglion (Fig. 852), which may be taken as a type of such collections, shows the entire ovoid mass to be enclosed by 2ifibrozis capsule continuous with that ensheathing the ner\'es. Immediately beneath the capsule the ganglion-cells are arranged in a fairly continuous layer of varying thick- ness, while the cells, more deeply placed, are broken up into groups by the tracts of Inter-fascicular septum Transverse section of small splenic nerve consisting chiefly of nonmeduUated fibres. X 200. ioo8 HUMAN ANATOMY. intervening nerve-fibres, a small amount of connective tissue prolonged from the endoneurium of the nerve-bundles and accompanying the blood-vessels being als4lia (ciliary, spheno-palatine, otic, and submaxillary), the i^^ani^lia within the three prevertebral j)le.\uses, and the innumerable small and often micro- scopic ganglia associated with the muscular tissue of the dii^estive, respiratory and uro-i^enital tracts, in the heart and in the various ylands. In their general structure the sym]Xithetic i^an^lia are similar to those connected with the spinal nerves, forming definite masses enclosed by a fibrous capsule, from which connective-tissue processes pass into the interior of the ganglion for the support and separation of the nervous elements. The individual gangli- Fic;. 85.). on-cells — unipolar, bipolar or multi- polar— are ensheathed by nucleated caj)sules continuous with the neuri- lemma of the nerve-fibres. The sympathetic ganglion-cells are \'ari- ously related to the terminal ramifi- cations of {a) other sympathetic neurones and of (i^) the neurones of the central nervous system (by way of the white rami fibres or their equi\-alents). In both cases, the ramification of the nonmedullated and fine fibre in the one and of the medullated fibre in the other, a pericellular plexus, commonly en- closes the cell-body. In the lower vertebrates (amphibians and reptiles), the spinal fibre frequently winds spirally around the single process of the ganglion-cell before breaking up into the pericellular plexus (Huber'). The broader relations of the component nervous elements of the spinal ganglia are considered in connection with the Sympathetic System (page 1354). Diagram of constiuieiits of spinal ganglion ; blue lines repre- sent efferent fibres ; black, afferent ; red, sympathetic ; a, sensory ganglion cells; c, cells of type II, whose axones end {b) around sensory cells; rf, sympathetic neurone; AR, PR, anterior and posterior roots; AD, PD, anterior and posterior primary divi- sions of spinal nerve; RC, ramus communicans. Fig. 855. DEVELOPMENT OF THE NERVOUS TISSUES. Refereace to the account of the early development of the nervous system (page 26) will recall the fact that the neural groove, later the neural tube, is lined by invaginated and thickened ectoblast from which the essential nervous tissues are derived. For the fundamental facts concerning the histo- genesis of these tissues we are in large measure indebted to the labors of His, whose account, supplemented by the important contributions of Kolliker, Cajal, Lenhossek, Schaper and others, forms the basis of our knowledge concerning these processes. Although in its principal features the histogenesis is similar in all parts of the neural tube, in that portion which becomes the spinal cord the changes are most typical and will, therefore, be here described. During the approximation and closure of the neural tube the ceils composing its wall undergo active prolife- ration, whereby the wall, at first composed of only one or two rows of definitely outlined cells, is converted into a multinucleated tract in which the cell boundaries dis- appear and the nuclei lie embedded within a general protoplasmic sheet or syncytium (Hardesty^). The large dividing elements within the latter, the fferminal cells of His, are conspicuous on account of their mitotic figures and are situated close to the lumen of the neural tube. His regarded them as special cells direcdy concerned in the production of the neurones, a conclusion, however, that has not 1 Journal of Morphology, 1S99. ^Amer. Journal of Anatomy, vol. iii., 1904. 64 ilm Segment from lateral wall of neural tube of pig embryo of 5 mm. ; syncytium replacing distinctly outlined cells, a, inner zone ; jt, germinal ctlls ; ibn, internal limiting mem- brane; ?«, peripheral zone ; r, radial strands of cytoplasm. '6^0. (Hardesty.) lOIO HUMAN ANATOMY. been sustained (Kolliker, Scha|>er and others) since the primary germinal cells probably only represent proliferating elements engaged in forming what for a time is an undiiTerentiated tissue. The cells composing the neural wall are at first in close contact, their blended cytoplasm (syncytium) forming an almost unbroken sheet. .Soon, liowever, this continuity is interrupted in consequence of the Iongiludin;il expansion of the tissue and the apjiearance of .spaces, and the cell-substance is resolved into a delicate reticulum, the myclospougium of His, which becomes condensed at the inner and outer margins of the wall of the neural tube into the internal and external limiting tnonbrane. The meshes of the reticulum enlarge, the intervening nucleated tracts of cytoplasm elongate and the increasing nuclei become radially disposed. Hy reason of these changes the elements ne.xt the lumen of the tube assume a columnar form and radial arrangement and become the priinarv cpendymal cells. The remaining elements, appropriately named the indifferent cells (Schaper ), increase in number in consecjuence of the continued division of the germinal cells and <^radually become collected as the nuclear layer at some distance beyond the ependymal zone. Meanwhile and very early, the peripheral portion of the supporting framework adjoining the outer border of the neural wall becomes denser and free from nuclei and is converted into the marginal zone ( Randschleier of Fig. S56. His), that is continuous with the r g Urn delicate reticulum pervading the otlier parts of the wall. The in- different cells later differentiate into ( a) the spongioblasts from which the characteristic constitu- ents of the definite supporting tissue, the neuroglia, are derived, and [b) the neuroblasts that are directly converted into the neu- rones. Within the resulting cell- complex that for a time occupies the greater part of the wall of the neural tube, it is difficult to distinguish with certainty between the neuroglia and neuron-producing elements, since both are often elongated in shape and prolonged into processes. Histogenesis of the Neuroglia. — In addition to the extension, conden.sation and moulding (by the developing ner\e-cells and fibres) that the primary syncytial mesh work undergoes elm Segment of wall of neural tube of pig embryo of 10 mm.; radial strands (r) of syncytium and differentiation of ependymal (at. nuclear (b) and marginal (»/) layers; Urn. elm, internal and external limiting membrane; ir. dividing cell ;/. pia mater. • 690. {Hardesty.) Fig. S57. Transverse section of ventro-lateral segment of . ; • .rd from pig embr>-o of 30 mm., upper part of figure from chrome-silver preparation, lower part from .n. : i . 1 wuii toluidin blue; f, central canal ; «'/>. ependymal layer; n. nuclear layer; »i, marginal layer; r, radial fibres; v, ventral plate uniting halves of cord. X 24°- (Hardesty.) (Hardesty), the gradual transformation of the spongioblasts and their descendants into fibrilla establishes a more definite framework that replaces the priman,- net-work (myelo.spongium), and eventually, in conjunction with the fibrilK-e derived from the processes of the ependymal cells, DEVELOPMENT OF THE NERVOUS TISSUES. lOll gives rise to tlie definite supporting tissue, the neuroglia. According to Hardesty, the gUa-fibres arise within the syncytial tissue independently of the neuroglia cells, a view in direct opjjosition to the observations of Ruhaschkin, who attriimtes to the descendants of the sjiongiohlasts, the gliagenetic cells, a jiositive rule in the production of the fibres. Accepting the conclusi(jiis of the last-named investigator, the successive stages of the cells concerned in the production of the general neurogliar tissue are represented by the spotii^ioblasts, the f^liof^cnetic cells, the aslroiyles, and, finally, the glia ct-lls. The primary cpcmtymal lifments are succeeded by the epithelium which lines the ventricles and the central canal of the spinal cord. Their perijih- erally directed processes are in large part transformed into glia-fil)res and thus, along with the processes of the spider cells, contribute to the formation of the neurogliar felt-work. The accompanying illustration (Fig. 857), taken from Hardesly's paper, afifords an instructive comparison of the appearance of the young supjwrting tissue after true staining with ai>proved reagents (Benda) and after silver jirecipitation methods (Golgi) upon which so mucii reliance has been placed. The silver picture shows the classic long neurogliar fibres e.xtending the entire thickness, but fails to reveal the wealth of supporting tissue and nuclei. To what extent the mesoblastic ingrowths that follow the penetrating young blood-vessels into the neural wall take part in the i>roduction of the distinctive neurogliar framework is admittedly difficult to determine (Hardesty) ; that such tissue, however, contributes to the support of the nervous elements is certain. Histogenesis of the Neurones. — The neuroblasts are distinguishable with certainty from the spongioblasts as soon as they are provided with nerve-processes. The latter appear cis out- growths from the pointed and peripherally directed ends of the developing nerve-cells, invade the marginal zone, and later emerge from the wall of the immature cord as the ventral or anterior root-fibres of the spinal nerves (Fig. 8581. The deeper tint of their distal ends after staining, their tendency to collect in con- verging groups, and the uniform width of the outgrowing nerve- processes are distinctive charac- teristics of the neuroblasts ( His ' ). The first, and for a considerable time the only processes with which the neurones are provided cor- respond to the axones that be- come the axis-cylinders of the efferent (motor) ner\-es. Subse- quently other processes, the den- drites, grow^ out in various direc- tions from the cell-bodies of the young neurones. Development of the Peripheral Nerves. — According to the teaching of His, accepted by most anatomists, the axis-cylinder of the entire future nerve-fibre is formed by the peripheral growth of the original nerve-process of the neuroblast. The assumed development of the nerve- fibre by the union of a numlaer of segments ( Balfour, Dohrn, and others, and, more recently, Bethe and O. Schultze) is not in accord with renewed investigations, and the findings upon which the composite theory of the fibre is based are open to different interpretation (Kolliker, Retzius). According to Bardeen,^ the development of the peripheral spinal nerves is briefly as follows: The motor neuroblasts and the sensory spinal ganglion-cells send out processes of considerable thickness, all of which soon begin to give rise at their extremities to groups of fibrillar, which increase in thickness and length and, in turn, at their extremities give rise to new groups of fibrils. At first the.se proceed as naked bundles, but soon become surrounded with sheath-cells of mesoblastic origin which thus enclose the early embryonic nerve, that may contain hundreds of fibrillse. After a nerve has become distended by ingrowth of new fibrils from behind, the proliferating sheath cells begin to wander from the periphery in among the fibrillar and give rise by anastomosis of their processes to a net-work that divides the original fasciculus into a number of secondary bundles. The intrafascicular cells increase rapidly, the process of subdivision 7-0- Xeurobhists Efferent axones Portion of spinal cord of human embryo, showing development of ventral root-axones as outgrowths from ventral neuroblasts. X 300. < After His. ) ' Die Entwickfclung des menschlichen Gehirns, 1904. 'Amer. Journal of Anatomy, vol. ii., 1903. IOI2 HUMAN ANATOMY. Fig. 859. '"*^^:- DevelopiiijBj iiitcicubtul nerve of pi^ enibr) o of 10 mm. ; lip of nerve is composecl of fibrils surroundt-d by shcalh-cells. ,-; 360. (Bardren.) Fui. - ••5^?* ^ continues and llie l)un(Jles of fibrillce l>econie progressively smaller and nujre compact until, surrounded by rnenil)raiious septa, they correspond to the axis-cylinders of the individual nerve- fibres, enclosed by tiie mnriUtntiia and its ceils. 'J'lie cudutnurium a|)|)ears comparatively late atid, like the neurilemma, is a product ot the mesoblast. Later, condensations of the mesoblast around the definite bundles of nerve-fibres and about the entire nerve-trunk provide \hk: pcii/iniriiiin and the cpiticurium respectively. Durinjj; its course t(^ the periphery the young nerve gives rise t(j numerous branches, the jjoints of outgrowth being indicated by a preparatory increase of the peripheral cells which often form a tubular projection into which the nerve- ribrilix* grow. The proximal plexuses (such as the brachial or lumliar) are formed during the outgrowth of the nerves from the region of the central nervous system ; the coarser distal plexuses arise during the extension of the branches to the various parts for which they are destined ; whilst the finer termiual plexuses are established during the development of functional unity between the nerve-fibres and the structures to which they are distributed. The medullary sheath is a comparatively late ac(|uisilion, since it does not appear until about the fourth month of hetal life. Within the central nervous system the tracts of nerve- fibres obtain their medullary coat at difTerent times (some not until after birth), a variation that is of much service in enabling the anatomist to trace the course of the individual paths of con- duction.' The origin and method of formation of the medullary substance has been, and in fact still is, a subject of di.scussion. it is, however, certain that its production is not de|)endent upon the neurilemma, since the medullated fibres within the cerebro-spinal axis are devoid of this sheath, and, further, that the myelin sometimes appears before the neurilemma (Kolster, Bardeen). While it is doubtful whether the myelin is directly formed from the outer part of the axis-cylinder, as suggested by Kolliker, it is probable that this structure exerts some influence resulting in the deposit of the myelin- droplets either from the blood (Wlassak), or from the apjiarently fluid substance that after a time surrounds the axis-cylinder (Bardeen). Regarding the formation of y.\\^ framework supporting the droplets of myelin, Hardesty' inclines to the view that certain sheath cells, which api:)ear during medullation, are probably ciincerned. From the foregoing account it is evident that the axis-cylinder is derived from the ectoblast and the neurilemma from the mesoblast ; the origin of the medullary sheath is still undetermined, but most probably is mesoblastic. Development of the Ganglia. — The origin of the afferent (.sensory) neurones, whose cell-bodies are situated within the spinal and other ganglia, is entirely different from that of the efferent (motor) ones above described. In the case of the spinal nerves, the develoi:)ment of the ganglia pro- ceeds from a group of ectoblastic cells that form a ridge, the p;atiglion-crest, on the margin of either lii^ of the still open neural tube (Fig. 860), just where the general ectoblast pa.sses into that lining the groove. On approximation of the lips of the latter, the cells of the ganglion-crests fuse into a wedge-shaped mass that completes the closure of the neural tube and constitutes a centre of proliferation from which the cells migrate outward over the dorsolateral wall of the tube. The proliferation is not uniform but most marked at points that correspond to the mesoblastic somites, in consequence of which a series of segmentally arranged cell-aggregations ajipears on each side of the neural tube. These collections are the aniages of the spinal ganglia. Within them certain cells soon become fusiform and, assuming the role of Tiieuroblasts, send out a process from either end. r)ne process — the axone— grows centrally, while the other — the dendrite — extends peripherally and bectmies the chief part of a sensory nerve-fibre. The subsequent growth of the neurone is not symmetrical, but to one side, and so D A - f- M Transverse sections of dorsal rcjrion of human embryos, sbowinK early differ- entiation of spinal KanK'io" ; A.li^ neural tube still open; C", D. tube closed; a, jfanKlion-ridRfs ; ^. fused ridges; c. out- growth to form jraiiRlion ; d, ectoblast. • 230. { l.enhossrk.) 'Amer. Journal of Anatomy, vol. iv.. 1905. DEVELOPMENT OF THK NERVOUS TISSUES. 1013 Fic. S61. dr^ m' ".■•*\:.: Cross-section of part of dorsal region of human embr>-o, showing developing spinal ganglion; dz, vz, mz, dorsal, ventral and marginal zones of spinal cord; dr, vr, dorsal and ventral root-fibres of spinal nerve (?;) ; sg, spinal ganglion on dorsal root. ■ 85. ordered that the two processes are approximated and finally joined to tlie cell-body by a common stalk ( Mj;. >S39). the neurone beinjj thus converted into an unipolar ganj;lion-cell. The centrally directed processes, the later posterior root-fibres of a spinal nerve, grow into the develop- ing cord and enter the peripheral zone (later the white matter) to end, when their development is completed, at various levels in relation with neu- rones formed within the neural axis. The peri- pherally directed jirocesses of the spinal sensory neurones, on the other hand, mingle with the axones from the motor neurt)nes to form the mixed nerves distributed to the various parts of the body. The essential parts of the sensory neurones, the cell-body and the processes, are derived from ectoblastic elements, whilst the sheaths, whether of the nerve-cells, of the fibres or of the entire ganglion, are contributed by the niesoblast. The development of the sympathetic ganglia, which include essentially three sets — those of the gangliated cords, tiiose of tlie prevertebral plexuses (cardiac, solar and hypogastric), and the terminal — has given rise to much discussion. According to one view, the sympathetic neurones have an independent origin and only secondarily form con- nections with the cerebro-spinal nerves. The other view, on the contrary, regards the sympathetic neurones as the direct descendents of neurogenetic elements derived from the developing spinal nerves. The evidence in support of the last view is so convincing that there is little question as to the correctness of its principle, although many details of the process, as relating to man, are still to be studied. It is, however, equally true that the sympathetic ganglia are neither produced by constriction and isolation of parts of the spinal ganglia, as sometimes assumed, nor by the migration of fully differentiated ganglion - cells, but, as emjihasized by Xeu- mayer, from undifferentiated neuroblasts which undergo in loco their development. The earliest suggestions of definite sympathetic ganglia in the human embryo appear about the beginning of the second fcetal month as aggregations of cells at the distal ends of the visceral rami of the developing spinal nerves. From these cells are derived the definite sympa- thetic neurones of the gangliated cord, as well as those which follow the mesial ingrowth of the spinal fibres for the pro- duction of the prevertebral and terminal ganglia. The lateral ganglia thus formed constitute for a time a series of isolated nodes ; subsequently these are connected by the differentiation of sympathetic axones which grow from one ganglion to the next and, in conjunction with the spinal fibres, establish the longitudinal commissural strands of the gangliated cord. Other sympathetic cells send axones centrally and give rise to the eflferent splanchnic nerves, whilst the axones of still others pass to the growing spinal nerves. Fig. 862. 5 %<. Spina gangliE ^~s>>i Sagittal section of rabbit embryo showing several developing spinal ganglia and nerve-trunks ; ^, aorta; S', intersegmental artery. X 52. IOI4 HUMAN ANATOMY. NER\K-TI:RM 1 X AT I ONS. The terminations of tin- ril)rcs composinjj: the j)cripheral nerves — the axones of certain motor neurones situated within the cerebro-spinal axis and the sympathetic system and the dendrites of the neurones of tlie sensory .tjani^dia — su|)ply the means by wliich tlie \arious structures of the body are brou.t;ht into intimate rehition with tlie nervous system. Some of these terminations transfer impulses resulting in muscular ct)ntractions ; t)thers convey impressions that produce \arious sensations (pain, pressure, muscle-sense, Fig. 863. temperature;. The nerve- Nerve ^ 4 terminations, therefore, may \ be grouped according to func- tion into motor and seyisory endings. Motor Nerve-Endings. The motor endings in- clude (a) terminations of the a.xones of neurones situated within the motor nuclei of the spinal cord and brain- stem that pass to voluntary muscle ; (^) terminations of sympathetic neurones that end in involuntary muscle and (r) in cardiac muscle. Endings in Voluntary Muscle. — On approaching their peripheral destination the medullated nerve-fibres branch repeatedly, each fibre in this manner coming into relation witii a number of mus- cle-fibres. When the med- ullated nerve-fibre reaches the muscle-fibre which it supplies, its medullary sheath abrui)tly ends and the neurilemma becomes inseparably fused with the sarcolemma, whilst the a.xis-cylinder passes beneath this sheath to terminate in an cnd-plale. The latter appears as an o\al area, from .040-. 060 mm. in its greatest diameter, which is applied to the muscle-substance ; in profile it shows a slight projection beyond the contour of the muscle-fibre, known as the eminence of Doycre. Embedded within a general nucleated sheet of granular proto])lasm, the sole-plate, lie the brush-like terminal arborizations of the axis-cylinder formed of irregular varicosites and club- shaped ends. From the details of the development of the motor end plates, as described by Rardeen, it is probable that the granular sole-plate and its nuclei are differentiated from the sarcoplasm and the nuclei of the muscle-fibre respectively. The much discussed relation of the end-plate to the sarcolemma — whether outside or beneath — seems to be decided in favor of a subsarco- lemmal position, since the muscle-sheath appears sub- sequendy to the formation of the motor-ending, a fact that explains the apparent piercing of the sarcolemma by the axis-cylinder. Usually each muscle-fibre is pro- vided with a single motor end-plate, which may lie at an equal or unequal distance from the ends of the fibre Motor nerve-endings in voluntary muscle ; bundle of ners'e-fibres is seen separating to supply the indi\idual muscle-fibres. X 160. Fig. 864. Motor nerve-ending in voluntary muscle; a, axone terminating in end- plate ; «, neurilemma ; j. sole-plate. - 400. Exceptionally two end-plates may be found on one muscle-fibre, in which case the endings lie near each other. NERVE-TERMINATIONS. 1015 Endings in Involuntary Muscle. — The terminations of the axones of the synipathctic neurones sujjplying the nunstriatcd muscle are comixirativeiy simple. The neurones contributing the immediate tibres of distribution usuaUv occupy the notial points of plexuses from which bundles Fio. S65. of nonmedullated nerve-tibres extend to and enclose the muscle • fasciculi. Entering the latter the nerve-tibres divide into delicate varicose threads that pass between the muscle-cells, jiarallel with their long axes. As they course within the intercellular substance, the varicose fibrils give of^ short lateral branches that end, as does also the i)arent fibre, in minute terminal knots on the surface of the muscle-cells, often in the vicinity of the nucleus. Probably by no means every muscle- cell individually receives a nerve-ending, a longitudinal group including three or four rows of muscle-cells lying between two adjoining terminal ner\e-fil)rils ( Huber). Endings in Cardiac Muscle.— These, also the termi- nations of svmpathetic neurones, have been studied bv, among -. ,. . . , 1 ^ •" 1 1-. • nil 1 II 1 ^ ' 1- Ncrve-endm,^ in iiivoluntarv others, Lajal, Retzms, Berkley and Huber. Accordnig to muscle. {Huber.) the last-named investigator, the_ varicose nerve-fibrils may be followed between the muscle-cells, during which course side branches arise that, as well as the main fibril, terminate on the muscle elements in endings of varying com- plexity. In some cases these are merely minute simple end-knots, resembling those found in involuntary muscle ; in other cases they are more elal^orate and consist of a group of secondary fibrilke bearing nodular endings, the whole recalling somewhat the motor end-j)lates in striped muscle. It is probable that most of the cardiac muscle-cells are in direct relation with nerve-endings (Huber). Sensory Nerve-Endings. Since the sensory endings are the peripheral terminal arborizations of the neurones whose cell-bodies lie in the spinal and other sensory ganglia, such teloden- dria are functionally the beginnings of the paths conducting the sensory stimuli to the central nervous system. According to their relations to the surrounding tissue, the sensory endings are broadlv grouped into free and encapsulated. Free Sensory Endings. — These endings include vast numbers of nerve- terminations found in the skin and the mucous membranes, chiefly within the epithelium but to some extent also wnthin the connective tissue strata. As a rule the sensory (afferent) nerve-fibres do not branch to any extent until near their peripheral destination, where they undergo repeated divisions, always at a node of Ranvier and in various directions. The medullary sheath of the main fibre is retained until close to' its termination, although some of its branches may course as nonmedullated fibres for a considerable distance before ending or entering the epithelium. In the skin — and the same general plan applies to the mucous mem- branes— the fibres destined for the epidermis lose their myelin coat beneath the basement membrane and enter the epithelium as vertically coursing nonmedullated fibrils. Within the epidermis they break up into numerous delicate fibrils which undergo further divi- sion into still finer varicose threads that ramify between the cells of the stratum germinativum and terminate in minute free end-knobs (Fig. 866). Although an intracellular position of these nerve- endings has been described by various writers, it is probable that the endings are extracellular and lie upon the surface of and not within the epithelial Free sensory endin-s within epidermis elements. Similar, but far less numcrous, free end- ?i/^''^','='" ^^T'^'lP'^^" "'^.'l'^-*^'"''"'*' ings, varicose and club-like in form, occur within terminate in end-knobs. {Dogtel.) , . • c ' i i • j u the connective tissue layers of the skm and the tunica propria of mucous membranes. Within the integument, conspicuous end- raniifications of sensorv neurones surround the hair follicles, lying upon the outer surface of the glassy membrane. ioi6 HUMAN ANATOMY. FK5. S67. Tactile cells of Merkel lying within inter- papillary ejiithelium ; broken line (<•) indicates junction of epithelium and connective tissue layer; (w) nerve passing into epithelium. X 160. ( lyorthmann.) Fig. 868. The tactile cells of Merkel, found in the ileei)er layers of the epidermis, represent a somewhat more chlTerentiated form of intraepithelial terminations and suggest transitions to the more specialized end- organs. In these endings the nerve-fibrils terminate in cu])-shaped e.xpansions or menisci, against which rest the modified ej)ithelial cells. The latter may be regarded as an imperfectly dinerentiated neurocpitlulium . e.xampics of which are seen in the gustatory cells in the taste buds and in the highly specialized visual and auditory cells in the retina and in the organ of Corti respectively. Encapsulated Sensory Endings. — In their most highly de\el()j)ed forms these end- ings (corptiscula nervorum terminaliaj are represented by relati\ely large special end- organs in which the terminations of the axis- cylinder are enclosed within an elaborate laminated capsule. The latter, however, is more often present as a much simpler and thinner envelope consisting of strands of fibrous tissue. Transition forms between the intraepithelial tactile cells above noted and the more specialized encapsulated end-organs, always within the connective tissue, arc seen in the corpuscles of Gra7idry (not found in man but cons])icuous in the skin covering the bill and in the tongue of many water-fowl), in which the nerve ends in a disc-like expansion enclosed between large modified epithelial cells and the neuromuscular and neurotendinous end-organs, presently to be described (page 1020). The group of simpler encapsulated endings includes three well-known examples : the e7id-bulbs and the genital corpuscles of Kraiise and the cor- puscles of Meissner, all of which possess a common structural plan — interwoxen telodendria embedded within a semifluid interfibrillar substance and surrounded bv a thin fibrous envelope. The End-Bulbs of Krause. — These endings include a variety of irregularly spherical or ellipsoidal bodies found in the edge of the eyelid, the conjunctiva and corneal margin, the lips and the oral mucous membrane, the glans penis and clitoridis and probably other parts of the integument highly endowed with sensibility. Within the conjunctiva, as tlescribed by Dogiel ', thev lie superficially jjlaced within the con- nective tissue near the summit of the papilhe and folds, when such elevations exist, but always close beneath the epithelium. They vary considerably in size, often being small (.002-004 mm.), but some- times measuring from .05-. 10 mm. in diameter. Usually a single nerve-fibre, exceptionally two or even more, enters each bulb, losing its medullary sheath as it pierces the thin fibrous capsule. Within the latter the nerve, now represented by the naked axis-cylinder, divides into from two to four branches, which, after describing several annular or spiral turns, give off varicose fibrils that undergo further division, the terminal threads forming a more or less intricate maze within the semifluid substance enclosed by the fibrous capsule. 'Archivf. mik. Anat., Bd. xliv.. 1895. ^^, '^^-.<\ ^,f Two corpuscles of Graiidry fron bill of duck ; nerve is seen entering corpuscle on right. X 265. Fig. 869. Two end-bulbs of Krause from human conjunctiva. (Dogiel.) NERVE-TKRM I NATIONS. 1017 Genital corpuscle from integument of penis; nerve divides before piercing capsule and terminates in intricate end- windings. (Dog^iet.) Genital corpuscle from integ- ument of human clitoris. X. 350. ( IVorlhniann.) The Genital Corpuscles. — Tliese endings, most numerous (from one to four to the s(|iKire niillimeler ) in the deeper strata of the corium coverinij the glans penis and cHtoridis, but occurring also in the neighboring parts of the genitalia, are of irregular oval or lobulated outline and from .02 to .35 ^'^'- ^""• mm. in diameter. They present the same general architecture as the end- bulbs, but are of larger size, possess a somewhat thicker capsule, and contain a more intricate interlacement of the terminal nerve-tibrilUe. The latter are derived from the subdivision of two or three meduUated tibres that enter near the base of the corpuscle and are beset with varicosities and club-shaped terminal enlargements. The fibrous capsule, consisting of several connective tissue lamellae possessing flat- tened fusiform nuclei, encloses the semifluid or granular interfibrillar substance in which the end-arborizations are embedded. The Corpuscles of Meissner. — In man these are most numerous in the corium of the skin covering the flexor surface of the fingers and toes. They are also found in other regions possessing sensibility in a high degree, such as the lips, margin of the eyelid, nipple, penis and clitoris, as well as on the dorsum of the hand and foot and the radial surface of the forearm. On the volar surface of the distal phalanx of the fingers, where they occur in greatest numbers, some twenty are found to the square millimeter (Meissner). The corpuscles occupy the summit of the papillae and ridges of the connective tissue stratum of the skin, and lie close beneath the cuticle, with their long axes perpendicular to the latter. In shape they are elongated irregular ellipsoids, often somewhat sinuous in outline, and in the larger papillae may be joined at the deeper end with others to form a compound corpuscle. They are relatively large, being from .12-. 18 mm. long and about one-third as wide. Depending upon the size, each corpuscle is sup- plied by one or more nerve-fibres which enter in the vicinity of the base, as the deeper end is called, and, on piercing the capsule and losing the medullary sheath, divide into a number of naked axis-cylinders. These pass across the corpuscle in parallel or spiral windings and are beset with fusiform and pyriform varicosities, similar enlargements marking the ends of the terminal threads. The entire fibrillar interlace- ment is embedded within a semifluid substance and enclosed bv a thin nucleated fibrous capsule. The Corpuscles of Ruffini. — These end- ings are also found within the skin, but at deeper levels, near and sometimes within the subcorium. They are of large size, sometimes measuring as much as 1.35 mm. in length, and of an elongated fusiform contour. The nerve-fibres, often two or more, which usually join the capsule on the side, less frequently near one end, retain the medullary sheath for some distance after penetrating the capsule and throughout e Fig. S72. ^ %^ S >i^ h>'' V f5| J Corpuscle of Meissner lying within papilla of corium of skin from finger; only deeper layers of overlying epidermis are shown; w, entering nerve-fibre. ■ 270. loi8 HUMAN ANATOMY. Ku;. .S73. Cylindrical end-bulb from con- nective tissue layer of skin. X iSo. ( Szymonowicz. ) a number of bold curves and twistin^s. Alter the disappearance of their sheaths, the naked a.xis-cvhnders undergo repeated divisions, the resulting fibrillar becoming varicose and intertwined and ending in free terminal knob-like enlargements. In contrast to the foregoing end-organs, in which the axis-cylinder subdivides into numerous terminal threads disposed as more or less elaborate intertwinings, a second group is flistiiiguished by the j)ossession of a thick laminated capsule that encloses a cylindrical core or inuey bulb containing the slightly branched axis-cylinder. These endings, of which the Pacinian corpuscle is repre- sentative, are relatively large and ellipsoidal. A transitional form, connecting them with the spherical end-bulbs, is presented by the cylindrical end-bulbs of Krause. These are found in \arious parts of the corium, the oral mucous membrane and between the bundles of striped muscle and of tendon. They are irregularly cylindrical in form, often more f>r less bent, and consist of a thin laminated capsule that encloses a core of semifluid substance in which lies the centrally placed axis-cylinder. The latter, after losing the medullary sheath on entering at the proximal end of the capsule, traverses the core without branching until near the distal pole, where it ends in a single or slightly subdi\ided terminal enlargement. The Vater-Pacinian Corpuscles. — These structures, the most highly special- ized sensory end-organs, are relatively large ellipsoidal bodies, from .5-1.5 mm. in length and about one-third as much in breadth, situated within the connective tissue in many parts of the body. In man they are found in f^'c. S74. the deeper layers of the connective tissue layer of the skin, especially on the palmar and plantar aspects of the fingers and toes, in the connective tissue in the vicinity of the joints, in tendons, in the sheath of muscles, in the periosteum and in the tunica propria of the serous membranes, the peritoneum, pleura and pericardium. They are particularly large in the mesentery of the cat. where they may be readily de- tected with the unaided eye as oval pearly bodies some- times two millimeters or more in length. The most conspicuous part of the Pacinian b^dy is the robust capsule that constitutes almost the en- tire bulk of the corpuscle and consists of from one to three dozen thin con- centric lamellae of fibrous tissue. The surfaces of the . lamellae are covered with endothelial plates whose nuclei appear as fusiform thicken- ings, along the concentric stria: of the corpuscle. The axis of the Pacinian body * i J J Valer-Pacinian corpuscles from skin of child's finger; A, lonertudinal; B, transverse section ; «. nerve entering capsule to reach inner bulb. ■ 1S5. NERVE-TF.RMINATK )XS. 1019 Fig. .S75. Corpuscles of Herbst from bill of duck ; a. longitudinal, *, transverse section ; n, nerve traversing lamellae of capsule ; axis-cylinder within core is surrounded by cells. X 360. is occupied by a corr or inner hulh of scmilliiiil substance in which the naked axis-cviinder is embedded. On joininj^ the pro.ximal pole of the corpuscle, the fibrous (Henle'sj sheath of the nerve-fibre blends with the outer lanielUe of the capsule, while the medullary coat is retained during the somewhat tortuous path of the fibre through the capsule as far as the core. Here the remaining envelope of the nerve-fibre disappears, the terminal part of its course, through the core, being as the naked a.xis-cylintler. At a variable distance but often just before gaining the distal pole of the core, the axis- cylinder divides into from two to four branches, each of which terminates in a slightly expanded end-knot. Some- times shortly after penetrat- ing the capsule, the nerve- fibre splits into two or more axis-cylinders which then share the common envelope of semifluid axial substance. Similar end-organs, the corpuscles of Herbst, occur in the velvety skin covering the bill and in the tongue of water-fowl. They closely resemble the Pacinian bodies of mammals, but differ in being generally smaller, relatively broader, and in exhibiting a row of cubical cells within the core and around the axis-cylinder. These cells are regarded as corres- ponding to the large cells enclosing the tactile discs in the Grandry's corpuscles. The Golgi-Mazzoni corpuscles, found in the subcutaneous tissue of the pulp of the fingers, are modifications of the ordinary Pacinian end-organs. They differ from the latter in possessing fewer lamellae, a relatively larger core and a more branched axis-cvlinder. Neuromuscular Endings. — First described by Kolliker and by Kiihne, although previously seen by Weissmann, these end-organs, often termed vmscle- spind/es, are now regarded as sensory endings that are probably concerned in afford- ing impressions as to tension or "muscle-sense". They lie within the connective tissue separating the bundles of voluntary muscle-fibres and are long spindle-shaped structures, varying in length from 1-5 mm. or more and in width from .1-3 mm. where broadest. They are widely distributed, being probably present in all the skeletal muscles, and are especially numerous in the small muscles of the hand and foot. They have not been found, however, in the intrinsic muscles of the tongue and in the eye muscles, although within the tendons of the latter very similar {neuro- tendmoics) end-organs ha\e been demonstrated. Each spindle consists of a capsule, composed of a half-dozen concentric layers of fibrous tissue, which encloses a group of usually from three to ten, but sometimes as many as twenty, striped muscle-fibres. meduUated nerves, blood-vessels and inter- spersed connective tissue. These intrafusal fibres, as they are called, differ from those of the surrounding muscle in being much smaller in diameter and length, markedly tapering towards either end, more coarsely but less distinctly striated, and in possessing nuclei within the sarcous substance. The striations are not equally distinct in all parts of the fibres, being much less evident in the middle zone than towards the ends. The fibres are more numerous and of greater diameter in the equatorial region than near the poles of the spindle. The intrafusal fibres collectively are surrounded by a thin special connective tissue envelope, the axial sheath, between which and the capsule lies the periaxial HUMAN ANATOMN' Capsule lymph-space. Each sj)iiidle receives usually several medullated nerve-fibres, which, after incorporation of their sheaths of Henle with the capsule, pierce the latter at various points and proceed to the individual muscle-fibres. The terminal relations of the nerves to the intrafusal fibres have been studied by means of the newer methods especially by Ruffini, ^ '■ "■• ^'^- Huber and DeWitt and Dokmc-I. After repeated division durinj^ their course throuifh the cajj- sule and periaxial space, the nerve-fibres pierce the axial sheath, lose their medullary coat and terminate either as one or more ribbon-like branches that encircle the mus- cle-fibres in annular or spiral windings, or, after further subdivision, as branched telo- dendria in which the ultimate fibrils end in irrej^ular spherical or pyriform enlargements. Neurotendinous End- ings.— These end-organs, described by Golgi and sub- sequently more fully investi- gated by Kitlliker, Ciaccio, and Huber and DeW'itt, in their general architecture resemble closely the sensory endings in muscle. They lie embedded within the interfascicular con- necti\'e tissue and are usually found in the vicinity of the junction of muscle and tendon. Like the neuromuscular end- ings, the taidon-spindles are long fusiform structures, from I. -1. 5 mm. in length, sur- rounded by a fibrous capsule. The latter encloses a group of from eight to twenty intrafusal tendon fasciculi, which are smaller and a])parently less mature than those of the sur- rounding tendon-tissue. The intrafusal fasciculi are invested by a fibrous axial sheath be- tween which and the capsule lies a periaxial lymph-space. On reaching the spindle, after repeated branching, the medullated nerve-fibres pene- trate the capsule, with which their fibrous ( Henle' s) sheaths blend, and undergo further division. The medullary coat is lost after they i)ierce the axial sheath, the naked axis- cylinders breaking up into smaller fibrils that extend along the intrafusal fasciculi. The terminal ramifications, applied to the surface of the fasciculi, vary in details (Huber). Some arise as short lateral branches that pardy encircle the fasciculi and end in irregular plate-like expansions, while others terminate between the smaller fasciculi. Axial sheath Nerve-fibre A, neuromuscular ending; B, neurotendinous ending in lonK'U'di- nal section, methylene-blue staining. ■ zfio. ( Drawn from preparation made by Professor Huber.) THE CENTRAL NERVOUS SYSTEM. TnK central iumaous system iiiclutles the s])inal coitl and tlie brain. In princi|jle tliL'Sc parts arc to l)c regarded as tlic walls of the primary neural tube, modified by unequal growth and expansion, which even after acquirini^ their definite relations enclose the remains of the canal, as represented by the system of ventricular spaces. In contrast to the spinal segment of the neural tube, which always remains a rela- ti\ely simple cylinder, the spinal corti, the cephalic segment early differentiates into three /!>/7w^?;-|' cerebral vesicles, the anterior and posterior of which subdivide, so that the secondary brain-vesicles are present. Coincidently marked flexure of the cephalic segment occurs at certain points and in consequence this part of the neural tube becomes bent upon itself to such a degree that the axis of the anterior vesicle lies almost i)arallel with that of the spinal segment (Fig. 91 2 j.. From the fi\e secondary divisions of the flexed and sinuously bent cephalic segment of the neural tube are developed the fundamental parts of the brain in the manner presently to be described (page 1060), whilst from the relatively straight spinal segment proceeds the development of the spinal cord, in which process growth and differentiation convert the originally thin-walled tube into an almost solid cylinder, the minute central canal alone remaining as the representative of the once conspicuous lumen. THE SPINAL CORD. The spinal cord (medulla spinalis) is that part of the central nervous system, or cerebro-spinal axis, which lies within the vertebral canal. Its upper limit, where it becomes continuous with the medulla oblongata, is in a measure conventional, since there is no demarcation on the cord itself to indicate exactly its junction with the brain. Accurately considered, the superior limit of the cord may be assumed to correspond with the emergence of the uppermost root-fibres of the first spinal nerve which pass out between the atlas and the skull ; this level also corresponds to the lowest strands of the pyramidal decussation of the medulla oblongata and to the upper border of the posterior arch of the atlas. For practical purposes, however, the lower margin of the foramen magnum defines with sufficient accuracy the upper limit of the spinal cord. Below, the spinal cord terminates somewhat abruptly in a pointed end, the cojius inediil/aris, that usually ends opposite the disc between the first and second lumbar vertebrae. The level to which the cord extends inferiorly, however, is subject to considerable \ariation, very rarely being as high as the middle of the body of the last thoracic vertebra ( Moorhead), or as low as the upper border of the body of the third lumbar vertebra (Waring). In the female subject the spinal cord, although absolutely shorter than in the male, extends to a relatively lower level in the vertebral canal. Slarked bending of the spine produces slight alterations in the position of the cord, during strong flexion an appreciable ascent of the lower end taking place. The relation of the cord to the vertebral canal varies at different periods. Until the third month of foetal life the cord occupies the entire length of the canal, but subsequently, owing to the more rapid lengthening of the spine than of the spinal cord, the latter no longer reaches to the lower limit of the canal and, therefore, apparently rises, so that by the sixth foetal month the lower end of the cord lies opposite the first sacral vertebra, and at birth terminates usually on a le\'el with the body of the third lumbar vertebra. Measured from its upper conventional limit to the lower end of the conus medullaris, the spinal cord in the adult male has an average length of 45 cm. [JjU in.), and in the female of 43.7 cm. {i-jM in.), in both sexes the proportion of the length of the cord to that of the pre- sacral spine being approximately as 64 : 100 (Ziehen). The cord-length bears no constant rela- tion to stature, although in a general wa^- tall individuals may possess long cords. The weight of the spinal cord, stripped of its membranes and nerves, is something less than 30 grammes (i oz. ), or about 1-2000 of the body-weight. Its proportion to the weight of the l)rain is r :43. When fresh the spinal cord possesses a soft cheesy consistence and a specific gravity of 1035. 1021 I02: HTM AN ANATOMY. Skull Pedicles, cut • Fig. S77. edulla I T 'if!i^^j I In a., c-^ Pedicles .XII T- Pedicles< ' LamiiKf, cut Iu«6P^^--~,Transverse vl" _..-^ •>«*, Anterior roots of spinal nerves _ Dura, ' reflected Spinal cord, covered with arachnoid and pia Upper part of spinal cord within dural sheath, which has been opened and turned aside; ligamenta denticulata and nerve-roots are shown as they pass outward to dura. median fibrous band, the septum posticiim, connects the posterior surface of the cord Fig. 879. Dural sheath Periosteum Spinal cord Posterior root Ligamentum deiiticulatum E.xtradural areolar tissue Anterior root Spinal ganglion Spinal nerve Vertebral artery . Body of fourth cervicil vertebra ^^ Transverse section of vertebral canal at level of fourth cervical vertebra, spinal cord in position. with the dura and partially subdivides the subarachnoid space. Lower, this partition, I024 HUMAN ANATOMY. Fit;. 880. Skiiii- Vcrlcbral at lory / i« ' Spinal '^' accessory nerve Pedicles, cut Mi-.lulla Spinal accessory ,. , , nerve Kdffe of cut dural sliealli Spinal tord I T //// Pedicles Edge ot cut dural shealh .Spinal cord I L Pedicles 5/« Posterior > divisions of / sacral nerves s: Cu - -End of conns medullaris Filuni lerminale Desiendinjt nerves End of dtiral shealh J Sll _ Iiliiin ixterniini — Cn End of fihnn Coccyx Posterior wall of vertebral canal has been removed and dural sheath opene/, I C. first cervical ner\-e and vertebra respectively; O/, coccygeal ner>'es. which may transmit blood-vessels, is imperfect or alto).jcthcr absent. As they (.TOSS the subarachnoid space the bundles of root-fibres of the spinal ner\es are enclosed by prolongations of the pia and arachnoid. These sheaths are retained by the nerves for only a short distance after the latter receive an additicjnal investment from the dura as they leave the vertebral canal. The dural sheath becomes continuous willi the epineurium of the spinal ner\es. The Cord - Segments. — Althoujj^h no sui^j^estion of such sub- division is to be seen as constrictions on its surface, in principle the sjjinal cord consists of a series of segments, each of which gives origin to the miterior (motor) and receives the/^.y- terior (sensory) root-fibres of one pair of spinal nerves. These nerves, usually thirty-one pairs in number, are classified as eight cervical, tweKe thoracic, fi\e lianbar, fixe sacral, and one coccygeal. Corresponding to the attachment of the nerves the cord is conventionally divided into cervical, thoracic, lumbar, and sacral regions. Of the entire length of a cord measur- ing 43 cm., approximately 10 cm., or about 23.5 i)er cent., belonged to the cervical region; 24 cm., or 55.5 per cent., to the thoracic; 6 cm., or 14 per cent., to the lumbar; and 3 cm., or 7 per cent., to the sacral region. The spinal nerves are attached to the lateral surfaces of the cord by fan-shaped groups of anterior and pos- terior root-fibres that are gathered into comjxict strands as they con\erge to form a common trunk (Fig. 884). The portifni of the spinal cord with which the root-fibres of a sjiinal nerve are connected constitutes its cord- segment, the limits of which lie in the inter\al separating the extreme fibres of the nerve and those of the adjacent nerves. In the thoracic cord these intervals are very evident, since the segments are relatively long ; in the cer\ ical and lumbar regions, on the contrary, the groups of root-fibres are so crowded that they form almost unbroken rows. The lenjjth of the individual cord- segments varies ; thus, according to the measurements of Liideritz, those of the cervical region, are from 11-13.5 mm. ; THI-: CENTRAL MCRX'OrS SVSTF.M. I02' those of the thoracic re- Kig. 88i. gion from 12-26 mm., tlie longest belonging to the V-VII thoracic nerves; those of the lunil)ar region rapidly decrease from 15.5 -5.5 cm., followed by a more gradual diminution to less than 4 cm. in tlie sacral region. In consecjiience of the disproportion between the lengtii of the spinal cord and that of the vertebral canal, the discrepancy be- tween the level at which the nerves are attached to the cord and that of the intervertebral foramina through which they leave the canal becomes more marked towards the lower end of the series. The growth of the cord, how- ever, is not uniform since, as shown by Pfitzner, during the later years of childhood elongation Transverse section of vertebral canal, at level of middle of first lumbar vertebra; spinal cord (conus medullaris), surrounded by nerve-bundles, is seen within dural sheath. Fig. 882. Conus Coccygeal I S medullaris Descending spinal nerves Front wall of dural sheath Filum internum of the thoracic region occurs to such an e.xtent that tills part of the cord once more e(iuals, if indeed not exceeds, the corresponding portion of the spine. While the cervical cord keeps fairly abreast the cer\ical portion of the vertebral column, the lumbar and sacral segments are left far behind. The results of these changes are seen in the course of the root-fibres, which in the neck, below the third nerve, run somewhat downward to their points of emergence, and in the thoracic region pass more horizontally, while those of the lumbar and sacral ner\'es descend almost vertically for a considerable distance — in the case of the last sacral nerve 28 cm. (Testut) — before reaching their appropriate levels. Coccvx End of spinal cord with roots of lower nerves descend ing in cauda equina to gain their respective foramina /-J In, 1-5 sn, en, lumbar, sacral and coccygeal nerves. The large and conspicuous leash of descending root-fibres, seen upon open- ing the dural sheath, constitutes the Cauda equina, in the midst of which the glistening silvery filum terminale is distinguishable. It is evident, there- fore, that in most cases the level of the cord-segment and that of the vertebra bearing the same designation do not correspond. Likewise, it must be re- membered that, although in general the spinal nerves are named in accordance with the vertebrae immediately below which they escape, in the neck there Filum are eight cervical spinal ner\'es and fnsSh' only seven vertebrae, the first or sub- cn occipital nerve emerging between the atlas and the skull, and the eighth between the last cervical and first thoracic vertebra; hence, e.xceptthe last one. they correspond with the vertebra below. 6.S I026 HIMAN ANATOMY. IlG Medulla f Cer\icai Thoracic LumV)ar Sacral Coccygeal Form of the Cord. — After rcm<)\al of its membranes and the root-fibres, the spinal cord is seen to dilifer from a simj)le cylinder in the following resj)ects. It is somewhat flattened in the antero-posterior direction, so that the sagittal diameter is always less than the transverse diameter, and its outline in cross-sections, therefore, is not circular but more or less oval ; its width is not uniform on account of two consi)icuous swellings that are associated with the origin and reception of the large nerves supplying the limJjs. The u|)per or cervical enlargement ( intumescentia cervicalis) begins just below the ujiper end ol the cord and ends opposite the second thoracic vertebra, having its greatest expansion at the level of the fifth and sixth cervical vertebrae, where the sagittal diameter is about 9 mm. and the trans\erse from 13-14 mm. The lower or lumbar enlargement ( intumescentia lumbalisj begins opposite the tenth thoracic vertebra, slightly above the origin of the first lumbar ner\e, and fades away in the conus medullaris below. It appears very gradually and reaches its maximum opposite the twelfth thoracic vertebra, where the cord has a sagittal diameter of 8. 5 mm. and a transverse diameter of from 11-13 ^^- (Ravenel). The lumbar enlargement is associated with the great ner\e-trunks supplying the lower limbs. The inter- vening part of the thoracic region is the smallest and most uniform portion of the cord and is almost circular in out- line. Where least expanded, opposite the middle of the thoracic spine, the cord measures 8 mm. in its sagittal and 10 mm. in its transverse diameter. These enlarge- ments appear coincidently with the formation of the limbs, are relatively small during fcetal life, and acquire their full dimensions only after the limbs have attained their definite growth. In a general way, a similar relation between the size of the enlargements and the degree of development of the limbs is observed in the lower animals. At the tip of the conus medullaris the spinal cord is prolonged into a delicate tapering strand, the filum terminale, that consists chiefly of fibrous tissue con- tinued from the pia mater and invested by arachnoid. It extends to the bottom of the pointed and closed end of the dural sac, which it pierces at the level of the second sacral vertebra and, ensheathed by a prolongation of dura (vagina terminalis), as the Jihim terminale externum, proceeds downward through the lower end of the sacral canal for a distance of about 8 cm. (3^ in.), finally to be attached to the periosteum covering the posterior surface of the coccyx. The part within the dural sac, the filum terminale intermnn. is about 16 cm. (61^ in.) in length and surrounded by the ner\ e-bundles of the Cauda equina (Fig. 882), from which it is readily dis- tinguished by its glistening silvery appearance. V Filum Spinal cord denuded of mem- branes and ner\-es. showing pro- portions of its length contributed by different regions and position and relative size of enlargements, as viewed from before; semidiagram- niatic. based on measurements ; one-third actual size. coccygeal nerves, homologous The upper half or less of the internal filum contains the terminal part of the central canal of the spinal cord walled by a thin and variable layer of ner\ous substance in which small ner\e-cells are usually present. The minute bundles of nerve- fibres often found adhering to the filum, which sometimes may be followed to and even through the dural sheath, are regarded by Rauber as representing one or two additional (second and third) with the caudal ner\-es of the lower animals. THE CENTRAL NERXOUS SYSTEM. ,027 The Columns of the Cord. -Inspection of the surface and particularly c,f cross-sections ot the spinal cord ( Fi^. H«5) shows the latter to be partially divided into a symmetrical rl^^du and left half by a median cleft in front and a partition in the mid-line behind. The cleti, the anterior median fissure ( fisstira mcdiana anterior j extends the entire len^nh of the cord, and is continued on the upper j.art of the filum terminale. It is narrow, from 2-3.5 '">"• in depth, penetrating^ for less than one-third of the ventro-dorsal diameter of the cord, and occupied by a process oi pia mater. Alonq; its Hoc.r, which lies immediately in front of the white commissure, it is frequendy deflected to one side of the mid-line and presents a slij.,du exi)ansion.' . Tlie separation into halves is completed by the posterior median septum (septum mediaiiuin postorius), the so-called posterior median fissure. With the e.\- ception of a shallovy groove in the upper cervical cord, the llimbar enlargement and the conus meduUaris, no fissure exists, but in its place a dense partiti(Mi extends from the posterior surface to the middle of the interior of the cord, entling inclose relation to the gray commissure. The character of the septum is a subject of dispute, according to some anatomists con- sisting exclusively of condensed neuroglia, while others regard it as composed of pial tissue blended with the neuroglia and, therefore, of both mesoblastic and ectoblastic origin. The latter view is substantiated by the mode of develojiment of the posterior septum, the immature pial covering of the developing l)lood-vessels being imprisoned within and hised with the neu- rogliar partition derived from the expanding dorsal halves of the developing cord (|>age 1050). The application of differential stains also demonstrates the composite nature of the septum. Each half of the spinal cord is further subdivided by the lines along which the root-fibres of the spinal nerves are attached. The root-line of the dorsal (sensory) fibres is relatively straight and narrow, and marked by a slight furrow, the postero- lateral sulcus (sulcus lateralis posterior) that lies from 2.5-3.5 "i""'- hiteral to the posterior septum and is evident even on the intersegmental intervals where the root- fibres are practically absent. The ventral root-line, marking the emergence of the anterior (motor) fibres, is much less certain, since the bundles of fibres of the indi- vidual nerves do not emerge in the same vertical plane, but overlie one another to some extent, so that each group occupies a crescentic area, whose greatest width cor- responds in a general way with that of the subjacent ventral horn of gray matter. The anterior root-line, which lies from 2-4 mm. lateral to the median fissure, is neither indicated by a distinct furrow nor con- tinuous. In this manner two longitudinal tracts, the posterior columns (funiculi posteriores) are marked off between the posterior median septum and the sulci of the posterior root- lines. These columns include something less than one-third of the circumference of the cord, and are about 6 mm. in width in the thoracic cordand 8 mm. and 7 mm. in the cervi- cal and lumbar enlarge- ments respectively. The tracts included between the dorsal and ventral root-lines constitute the lateral columns (funiculi laterales) and those between the ventral root-lines and anterior median fissure are the anterior columns (funiculi anteriores). Such subdivision into anterior and lateral Fig. 884. Medulla 4 cerv Ganglion on 4 nerve Dorsal nwts of s cerv. nerA'e Upper end of spinal cord, viewed from behind after partial removal of dural sheath ; cord-segments are indicated by groups of converging bundles of posterior root-fibres; spinal ganglia are seen lying within the intervertebral foramina; spina! accessory nerve is seen ascending on each side. I028 Hl'MAX ANATo.MV columns is, however, largely ariilicial, since neither sui«rficially nor internally is there a tlefinite deiiiarcation between these tracts. They may be, therefore, conveniently regar^led as forming a common antcro-lateral column, that on each side embraces something more than two-thirds of the semicircumference of the cord. In the lower cervical and upper thoracic cord, each posterior c- variable and uncertain owing to the changes incident to the use of hardening fluids. In a general way when well preserved the lumen is round or oval and smallest in the thoracic region ; in some places, as in the upper cervical cord and in the lumbar enlargement, it is larger and often appears pentagonal in outline, whilst in others the calibre may be reduced to a sagittal slit. The position of the central canal varies at difTerent levels in relation to the ventral and dorsal surfaces of the cord. In the middle of the lumbar region it occupies approximately the centre of the cord, but above, in the thoracic and cervical segments, it lies much nearer the ventral than the dorsal surface, while below it gradually approaches the dorsal surface, but always remains closed. Mention may be made of a remarkable structure named Rcissjier''s Jibrc, after its discov- erer, that as a longitudinal thread of great delicacy lies free within the central canal of the cord and the lower ventricle of the brain, extending from the cavity of the mesencei^halon above to the lowest part of the cord-canal below. The interpretation of this structure as an artefact, which considering its extraordinary position is most natural, seems untenable in view of the positive testimony, confirming its existence as a preformed and true structure in many vertebrates, given by several subsequent observers and especially by Sargent.' Its nature and significance are problematic. Although the existence of this fibre has been established in many vertebrates, even in birds, it has not yet been discovered in man. MICROSCOPICAL STRUCTURE OF THE SPINAL CORD. The three chief components of the spinal cord — the ner\e-cclls, the nerve-fibres and the neuroglia — vary in jjroportion and dis|)osition in the white and gray matter. It is, therefore, desiral)le to consider the general structure of the cord before describ- ing its detailed characteristics at different levels. The Gray Matter. — The most distinctive elements of the gray matter are the vnillifyolar nenostero- lateral f^roup. Below the level of the second thoracic nerve the dorso-lateral cell-column is unrepresented as far as the second sacral segment where it reappears, somewhat abruptly, and attains its maximum size in the fourth and fifth lumbar segments. The column then diminishes and ceases at the lower part of the third sacral seg- ment. Within the sacral cord, between the levels of the first and third nerve inclusive, the dorso-lateral cell-group is augmented by an accessory group. From the third lumbar to the sacral nerve-levels, an additional compact collection of nerve-cells occupies a more median position in the anterior horn and constitutes the central group. From the position of the greatest expansions of the lateral cell-columns— within the cer\-ical and lumbo-sacral enlargements — it is evident that they are associated with the large nerves sup- plying the muscles of the limbs. Further, according to Bruce, in a general way the size of the radicular cells bears a relation to that of the muscles supplied, the smaller dimensions of the cervical cells, as compared with those of the lumbo-sacral region, corresponding with the smaller size of the upper limb in comparison with that of the lower one. In addition to the nerve-cells assembled within the foregoing more or less well defined groups, some scattered cells are irregularly distributed through the anterior horn and do not strictly belong to any of the groups. Below the level of the first coccygeal nerve, the cells of the anterior horn become so diminished in number, that they are no longer grouped with regularity, but, reduced in size, lie uncertainly distributed within the gray matter as far as the lower limits of the conus medullaris. The nerve-cells of the posterior horn are neither as large nor as regularly disposed as the anterior horn cells. Only in one locality, along the median border of the base of the posterior horn, are they collected into a distinct tract, the colimin of Clarke ; otherwise they are scattered without order throughout the gray matter of the posterior cornu. Since, however, the latter comprises certain areas, the cells of the posterior horn may be divided into (i) the cells of Clarke s column, (2) the cells of the substantia gelatinosa Rolandi, and (3) the inner cells of the caput cormi. The cells of Clarke's column form a very conspicuous collection which extends from the lev'el of the seventh cervical nerve to that of the second lumbar nerve and is best developed in the lower thoracic region of the cord. Although confined chiefly to the dorsal portion of the cord, and hence sometimes designated as the "dorsal nucleus," Clarke's column is represented to a slight degree in the sacral and upper cervical regions {sacral and cervical nuclei of Stilling) . In cross-sections the cell-column appears as a group of multipolar cells that occupy the mesial border of the base of the posterior horn and, where the column is best developed (opposite the origin of the twelfth thoracic nerve), correspond to an elevation on the surface of the gray matter. The cells usually are about .050 mm. in diameter, polygonal in outline and possess a relatively large number of richly branched dendrites that radiate chiefly within the limits of the group (Cajal). The axones commonly spring from the anterior or lateral margin of the cells and course ventrally for a considerable distance before bending outward toward the lateral column of white matter within which, as constituent fibres of the direct cerebellar tract (page 1044) , they turn brainward. 'Topographical Atlas of the Spinal Cord, 1901. I034 HIMAN ANAl )M\ . The nervc-cells of the substantia gelatinosa Rolandi, also known as Gierke s cells, include innumerable small stellate, less tre(|uenlly fusiiurm or i>e ir-shaped elements that measure only from .U06-.020 mm., althou.ich exceptionally ol l.irger size. I'heir numerous short dendrites are irre^^ularly disposed and !jrancheed or pyramidal in form, their long axes lying parallel or the apices directed towards the Rolandic substance respectively, and constitute a one-celled layer enclosing the substantia gelatinosa. into which many of their tangentially coursing dendrites penetrate. Their axones pass through the substantia gelatinosa and probably continue for the most part within the lateral column, although some enter the posterior column (Cajal, Kolliker). The inner cells of the posterior horn are intermingled with numerous nervous elements of small size irregularly distributed within the head of the dorsal cornu. The inner cells proper are triangular or spindle-shaped in form and, on an average, measure about .050 mm.; they are, therefore, larger than the ordinary cells of the Rolandic substance. The dendrites arise Fig. 8S9. White mattei of -.-.,' ■ ' '^ ^ ~ ~ ^ . j^^, posterior column J!2iL-M:^i.J!^Z~.... ^_ \ _ V'-'-iv^ / - ' i , /• • . ''^:'»^V*' ^■■'-^■^•^'■-^: ■■. .^'> ■■■;■■ ^v'-' ^r'- ?-- ■;,;■■•..■.*■■ .-. .% ■ ■ ^^\->*:^-:. -^y.:^ ir^r^-'- Cells of Clarke's column ^^^^f^'^iy^^dy ''" %^^^^ ^ Substantia gelatinosa centralis ^ei.-' -•.^, /^^ t.f'' Central canal Part of cross-section of cord, showing cells of Clarke's column in base of posterior horn. X no. from the angles or ends of the cells and diverge in all directions. The axones pass, either directly or in curves, mostly into the lateral column of the same side ; some, however, have been followed into the jwsterior or anterior columns of the same side (Kolliker), and, rarely, into the opposite anterior column (Cajal). Exceptionally type II cells— those in which the axone is not prolonged as the axis-cylinder of a nerve-fibre, but soon breaks up into an elaborate end- arborization confined to the gray matter— are found within the gray matter of the posterior horn. Their number is, however, much less than often assumed (Ziehen). The nervous character of most of the cells seen within the substantia gelatinosa Rolandi has been established only since the introduction of the Golgi methods of silver-impregnation. Previously, these elements were regarded as glia cells, an exceptionally large amount of neuroglia in general being attributed to the Rolandic substance. It is now admitted that instead of such being the case, this region of the gray matter is relatively poor in neurogliar elements and numerically rich in nerve-cells. The nerve-cells of the pars intermediate of the gray matter, which connects the dorsal and \entral horns ami lies opposite the gray commissure, may be broadly divided into two classes, the lateral and the middle cells, that occupy respectively the outer border and the more central area of this part of the grav matter of the cord. MICROSCOPICAL STRUCTURK OF SPINAL CORO. 1035 Those of tlic fust class, or intermedio-lateral cells, are associated witli the forniatio reticu- laris and its condensation, the lateral horn, and lience are often spoken of as the t^roup or column of Ihc lateral horn. These ceils form a slender tract of small closely packed elements that is represented throu.uii almost the entire len^nh of the cord, althoujih best marked in the upper third of the thoracic re<;ion and i)artially interrupted in the cervical and lumbo-sacral segments. Where the formatio reticularis is condensed with a distinct lateral lujrn, as in the thoracic region, the cells occupy the projection, but elsewhere lie within the base of the gray net- work. As a continuous cell-cohunn the tract extends from the lower part of the eighth cervical segment to the upper i)art of the third lumbar, being most conspicuous at the level of the third and fourth thoracic nerves (Bruce). I'ractically su|)pressed in the cervical region between the eighth and third segments, above the latter the column reappears along with the formatio reticularis. Below, it is again seen within the third and fourth sacral segments. The nerve-cells are multi- polar or fusiform in outline, from .015-045 nun. in their longest diameter, contain little pigment, and are provided with a variable number of dendrites, of which two are usually larger than the others. These ari.se from opposite poles of the cell and send brandies, for the most part, into the adjacent white matter. The axones pass directly into the lateral columns and become ascending or descending tit^res ; a few a.xones, however, enter the anterior column of the same side (Ziehen). The cells of the second class, or intermediate cells, are irregularly disjwsed and only in the upper part of the cord present a fairly distinct middle group (W'aldeyer). They are polygcMial or fusiform in outline, small in size (seldom exceeding .025 mm.) and provided with irregular dendrites. The axones are continued chiefly within the lateral column of the same side, altlunigh some pass to the anterior column and a few probably cross to the opposite side. A small number of isolated nerve-cells are usually to be found within the white matter, out- side but in the neighborhood of the gray core. These, the outlying cells of Sherrington,' by whom they have been studied, occur most frequently in the vicinity of the more sui)erficially placed cell-columns. Within the anterior columns they lie in the paths of the fibres proceeding to the anterior white commissure ; in the lateral columns they are in proximity to the intermedio- lateral group of the lateral horn and formatio reticularis and to the cells of the substantia Rolandi ; and in the posterior columns, where they are relatively numerous, they are associated with the fibre-tracts leading to the column of Clarke. The outlying cells are regarded as elements displaced from their usual position during the course of the differentiation and growth of the white and gray matter. .Similar displacement sometimes aflfects the cells of the spinal ganglia, which then may be encountered within the cord. The Neuroglia of the Gray Matter. — As in other parts of the cord, so in the gray matter the neurogHa is everywhere present as the supporting framework of the nervous elements, the cells and fibres. The gen- ^"^- ^''9'^- eral structureof neuroglia having been described (page 1004), it only re- mains to note here the special features of its arrangement within the gray matter. In general, the felt-work of the neu- rogliar fibrils is more compact than that per- meating the white matter, being somewhat denser at the periphery than in the deeper parts of the gray matter. There is, however, no hard boun- dary between the sup- porting tissue of the two, since numerousglia fibrils extend outward from the frame-work of the gray matter to be lost between the nerve fibres of the adjoining columns. This feature is marked in the anterior horn, where the glia fibrils form septa of considerable thickness that diverge into the surrounding columns ; further ^ Proceedings Royal Society, vol. 30, 1890. Posterior median septum Paramedian subdividing posterior column eptum Anterior median fissi*-e Anterior column Transverse section of cord slightly magnified, showing general arrangement of neuroglia. X lo. I036 HUMAN ANATo.MV. the conspicuous processes of the iorniaiio reticularis aiul the projecting lateral horn consist largely of neuroglia. The larger nerve-cells and their robust processes are ensheathed by interlacements of neuroglia tibrilla-. In the several parts of the posterior horn the amount of neuroglia \aries. Thus, tin- ai)ex consists alnutst exclusively of glia tissue, while within the Rolandic substance the number of glia fibres and cells is unusually small. Within the caput and remaining jjarts of the posterior horn the neurogliar elements are similar in quantity and disposition to those in the anterior horn. Thf cpendyma cells lining the central canal of tlie cord are the direct descendants of the radially arran<;ed enihryonal supporting elements (page 1004) ; they may, therefore, be regarded as sjHfcialized neuroglia cells. Althougli most advantageously studied in the f(i.tus and the chiUi, in favorable preparations from adult ct>rds tliey are seen as a single row of pyramidal cells, from .030-.050 nun. long and from one-fourth to one-third as broad, who.se bases are directed towards the lumen of the canal and beset with cilia. Their pointed distal ends, or apices, are prolonged into a long delicate cpeudymal fibre^ that in the adult is soon lost in the surrounding neuroglia, but in the fatus extends through the entire thickness of the cord. The ependyma cells are not all of equal size, those occupying the ventral mid-line, especially in the cer\ical region, being about twice as long as those on the opposite wall of the canal. The ejien- dymal fibres proceeding from these cells are of special length and thickness, the ventral ones con- verging to form a wedge-shaped mass that in the young subject continues as far forward as the bottom of the anterior median fissure. The dorsal ependymal fibres are prolonged through the gray commissure into the posterior median septum, some diverging into the columns of Goll. Substantia gelatinosa centralis is the name given to a zone of peculiar trans- iucency that immediately surrounds the central canal. This annular area consists of modified neuroglia in which radial ependymal fibers are F'f''- %'• interwoven with circularly disposed neurogliar fibrillae, -rS:i23e^>;o_ the whole giving rise to a compact stratum, interspersed ■ "'V^f ^ with an unusual number of glia cells, upon which arrange- ,, r.v \ ment, in conjunction with the absence of ner\e-fibres, ,''■'■ the characteristic appearance of the gelatinous substance I* \ depends. In addition to the branched glia elements, a ' . ' number of radially directed spindle cells are present in this zone ; they send delicate processes between the ' ■;'.' ependyma cells, of which they are probably outwardly ' -^ displaced members. In marked contrast with the Ro- ;' ■ , landic substance, which caps the posterior horn, the . ; ^ ,' \ substantia gelatinosa centralis contains only a few small . ": • , ^- nervous elements, in recognition of which the term, sub- ^y stariiia ^liosa coifralis, has been proposed bv Ziehen. The Nerve-Fibres of the Gray Matter. — ' : } ' Within all portions of the gray core a considerable part ,--~ of the intricate ground-work in which the nerve-cells lie "-— ^- embedded is contributed bv the processes of neurones subs^am\?VeTa"tTnotl'«nt;a'f;rf^^^^^ situated at the Same, different or even remote levels. child's cord ; canal is lined with Thcse proces.ses, which coustitutc the ncpve-fibres, ependvma cells, outside of which 1 11 . j 1 1 11 1 .1 lies neuroglia with glia cells. ,• 135. mcdullated and uonmeduUated, that are seen traversmg the gray matter in all directions, include: ( i ) the collate- rals and the terminal branches of the dorsal root-fibres that enter the gray matter ; (2) ner\e-fibres of the descending tracts that terminate in relation with the ventral (motor) horn cells ; (3) the axones and collaterals given off by the numerous pos- terior horn cells, that traverse the gray matter t() and from the respective columns into which they pass. The dendritic j^necesses, as well as the axones of the tvpe II cells, also contribute to the sum of nervotis hbrilke encountered within the gray matter of the cord. WHITE MATTER OF THE SPINAL CORD. The predominating components of the white substance being the longitudinal ner\e-hbres which pa.ss for a longer r»r shorter distance up and down in the columns of the cord, in cross-sections the outer held, between the gray core and the periphery Wlini': MATTKR OF THE SPINAL CORD. 1^37 of the cord, appears to be composetl of iniuunerable, closely set, small cells, held toii^ether by delicate supportniii;- tissue. These ai)j)arent cells are the niedullatcd nerve- fibres cut transversely, in which the sectioned axis-cylinders show as deeply stained dots, that commonly lie somewhat eccentrically and are surrounded by deli- cate irrei^ularly annular striations representing the framework of the medullary c; 230. having an extended course being larger than shorter ones ; it follows that the fibres occupying the peripheral parts of the white matter, particularly in the lateral columns, are more frequently of large diameter than those near the gray matter. The immediate surface of the white substance beneath the pia mater is formed by a con- densed tract of neuroglia, the subpial layer, from .020-.040 mm. in thickness, that is devoid of nervous elements and forms the definite outer boundary of the cord. This zone consists of a dense interlacement of circular, longitudinal and radial neuroglia fibrils among which numer- ous glia cells are embedded. From the deeper surface of this ensheathing layer numerous bundles of fibrilke penetrate between the subjacent nerve-fibres to become lost in the general supporting ground-work. At certain places the bundles are replaced by robust septa by which the nerve-fibres are imperfectly divided into groups or tracts, as conspicuously seen in the pos- terior column where the paramedian septum effects an imperfect subdivision into the tract of Goll and of Burdach. The blood-vessels that enter the ner\-ous substance from the pia, accom- panied by connective tissue, are surrounded by tubular sheaths of neuroglia, and the same is 1038 Hl'MAX ANAIOMV. true of tile bundles of root-fibres of the spinal nerves. Hut a|)art from the connective tissue that enters with the blood-vessels, the amount of mesoblastic tissue concerned in the supporting framework of the cord is inconsiderable, according to some histologists, indeed, being practically nothing. Fibre-Tracts of the White Matter. — Althou<;h microscopical examination of ordinary sections of the cord affonls slij.,dit indication of a subdivision of the columns of white matter into areas corresj)onciing^ with definite fibre-tracts, yet the combined evidence of anatomical, patholot^ical, embryolog^ical and experimental in\esti.!L,'^ation establishes the existence of a number of such jjaths of conduction. With few exceptions, they are, however, without sharp boundaries and illy defined, adjoining tracts often overlapping, and depend for their presence upon the fact that nerve-fibres ha\ing the same function and destination proceed in coiii]jany from the same grouj) oi ner\e-cells ( nucleus ) along a similar course. In addition to being pro- vided with paths of conduction necessary for the performance of its function as a centre for indei)endent (reflex) imjjulses in response to external stimuli, the cord contains tracts that connect it with the brain, as well as those that bring the various levels of the cord itself into association. The white matter, therefore, contains three classes of fibres : ( i ) those entering the cord from the periphery and other parts of the body ; (2) those entering it from the brain ; and (3) those arising from the ner\e-cells situated within the cord itself. The first two constitute the exogenous, the last the endoi^cnous tracts. It is evident that some of these fibres constitute pathways for the transmission of impulses from lower to higher levels and hence form ascaidhig tracts, while others, which conduct imjjulses in the opposite direction, form descending tracts. Since it is impossible to distinguish between these fibres by mere inspection of sections of the adult normal cord, and, moreover, extremely difficult and practically impossible to follow in such preparations the longer fibres throughout their course, advantage is taken of other means V)y which differentiation of individual tracts is feasible. Such means include chiefly the e.xperimental and embryological methods. The experimental method depends upon the law discovered by \\'aller, more than half a centurj- ago, that when the continuity of a nerve-fibre is destroyed, either by a pathological lesion or by the experimenter's knife, the portion of the nerve-fibre (the axone of a neurone) beyond the break, and therefore isolated from the presiding nerve-cell, undergoes secondar>' degeneration, while the portion remaining connected with the cell usually undergoes little or no change. It should be pointed out, however, that occasionally the connected portion of the fibre, and even the nerve-cell itself, undoubtedly exhibits changes known as retrograde degen- eration, which, although uncertain as to occurrence and cause, may at times prove a source of error in deducing conclusions. If a lateral section of one-half of the cord of a li\ing animal be made, and, after the expiration of from three to four weeks, transverse sections be cut and ap])ropriate1y prepared (by the methods of Marschi or of Weigert), certain groups of nerve- fibres will present degenerative changes. It will be seen, however, that the degenerated tracts in sections taken from above the lesion are not the same as those in sections from below the division, showing that certain fibres have been involved in opposite directions, those arising from nerve-cells lying below the lesion being affected with ascending degeneration, and those from cells situated above with descending degeneration. In this manner, by careful study of consecutive sections, much valuable information has been gained as to the origin, course, ter- mination and function of many fibre-tracts within the central nervous system. The embryological method, also productive of important advances in our knowledge of the nervous jiathways, is based on the fact, first demonstrated by Meckel, that the nerve-fibres of the central nervous system do not all accjuire their medullary sheath at the same time. Taking advantage of such variation, as suggested by Meynert and later extensively carried out bv I'lechsig and other.^, upon staining sections of embr}onal tissue with reagents that color especially the medullary substance, it is possible to differentiate and follow certain fibre-tracts in the foital cord with great clearness, since only those tracts are stained in which the myelin is already formed. It is of interest to note that, in a general way, the order in which the different strands of the cord accjuire their medullary coat accords with the sequence in which nervous function is assumed by the fretus and child. Thus, the paths required for spinal reflexes (the posterior and anterior root-fibres) are first to become medullated (fourth and fifth ftetal months); those bringing into association the different segments of the cord next (from the fifth to the seventh month) acquire myelin ; those connecting the cord with the cerebellum follow somewhat later, while those establishing relations with the cerebral cortex are last and do not begin to medullate until shortiv before birth. WIIITI'. MATTER OF TlIF SIMXAL (^ORD. 1039 Based on tlic collective evidence conlribuled by these methods — anatomical, physiological, and developmental — it is possible to locate and trace with fair accuracy a number of tibre-tracts in the cerebro-spinal axis. Since they are undergoing continual augmentation or decrease, their actual area and jiosition are sul)ject to variation, so that the detailed relations in one region of the cord differ from those at other levels. The accompanying schematic figure, therefore, must be regarded as showing only the general relations of the most imjiortant paths of the cord, and not as accurately rei)resenling the actual form and size of the fibre-tracts. It must also be api)reciated that the definite limits of these tracts in such diagrammatic Fig. S93. 1 Association tracts Kubro-spinal tract Vestibulo-spitial tract '- Spino-thalamic tract Spino-o!ivary tract (Helweg) Vestibulospinal tract Cerebro-spinal tract ~Tecto-spinal tract Diagram of spinal cord, showing position of chief tracts and relations of their component fibres to nerve-cells; 1-5, posterior root-fibres entering root-zone (R.Z.) and Lissauer's tract (L.), open circles (o) indicate that fibres pass up and down ; c, c, collaterals from long ascending tracts (1,2) to anterior root-cells; 3, fibres ending around cells of Clarke's column; 6, fibres forming direct cerebellar tract; 7, 8, fibres forming (iowers' tract; 9, 10, fibres from lateral and direct pyramidal tracts; 11, 11, anterior root-fibres; V.F., ventral field; O.F., oval field; C.B., comma bundle. representations seldom exist in reality, since the fibres of the adjacent paths in most cases overlap, or, indeed, extensively intermingle, so that the fields seen in cross-sections may be shared by strands belonging to different fibre-systems. The Fibre-Tracts of the Posterior Column. — The subdivision of the posterior column of white matter by the paramedian septum into two general parts has been noted (page 1028). Of these the inner one is the postero-median fasciculus, or tract of Goll (fasciculus gracilis), and the outer one is the postero- lateral fasciculus or tract of Burdach (fasciculus cuneatus). These tracts are so intimately associated with the fibres entering by the posterior roots of the spinal nerves, that the general relations and behavior of these fibres must be considered in order to understand the composition of the posterior columns, as w-ell as that of certain secondary paths. All sensory impulses that enter the spinal cord do so by way of the posterior root-fibres. The latter are the centrally directed processes (axones) of the neurones whose cell-bodies lie within the spinal ganglia situated on the dorsal roots of the spinal nerves. They convey to the cord the various impulses collected by the peripherally directed processes (the sensory nerves) from the integument, mucous membranes, muscles, tendons and joints from all parts of the body, with the exception of those served by the cranial nerves. The impulses thus conducted are transformed into the impressions of touch, muscle-sense, heat, cold and pain. The last being probably the result of excessive stimulation that by its intensity causes discomifort in various degrees, the existence of special paths for the conduc- tion of painful impressions is unlikely. It is evident that the larger part of the 1040 lllMAX ANA'IOMV Fig. .S94. sensory neurones lies outside the spinal ccjrcl ; it is. however, with the intramedullary portion of these neurones, as constituents of paths within the cord, that we are here concerned. On entering; the spinal cord aioiij; the posterolateral j^rooN e. the dorsal root- fibres for ihr most part penetrate the tract of Burdach, close to the inner side oi the posterior horn. Some of the more external root-tihres. however, do not. enter Hur- dach's tract, but form a small adjoinin^^ held, the tract of Lissauer, that lies im- mediately dorsal to the yi)e.\ of the posterior horn. Soon after gaining the posterior column, with few exceptions, each dorsal root-fibre undergoes a > or |— like divi- sion into an ascending and a descending limb, which assume a longitudinal course and pass upwartl and downward in the cord for a variable distance, the descending limb being usualh- the shorter. During their course from both, l)ut particularly from the descending limb and from the proximal part of the ascending fibre, collateral branches are given off which bend sharply inward and pass horizontally into the gray matter to end chiefly in relation with the neurones of the posterior horn, from which cells secondary j)aths arise. Not only the collaterals, but also the main stem-fibres of the descending and shorter ascending limbs end in the manner just described. In addi- tion to the short collaterals destined for the cells of the dorsal horn, others, the ventral reflex collaterals, pursue a sigmoid course, traversing the substantia gelatinosa Rolandi and the remaining parts of the posterior horn and the intermediate gray matter, to end in arborizations around the radicular cells of the anterior horn, and thus complete important reflex arcs, by which impulses transmitted through the dorsal roots directly impress the motor neurones. The latter are usually of the same side, but some collaterals cross by way of the anterior commissure to terminate in relation with the anterior horn cells of the opposite side. It is probable that a considerable number of such anterior horn reflex collaterals are given off from the fibres that ascend in the long tracts of the [wisterior column to the medulla oblongata. With possibly the exception of certain fibres which pass directly to the cerebellum (Hoche). all the sensory root-fibres ( axones of neurones of the I order) end around the neurones situated either within the gray matter of the spinal cord or within the nuclei of the medulla ; thence the impressions are conveyed by the axones of these neurones of the II order to higher centers, to be taken up, in turn, by neurones of the III or even higher order, in the sequence of the chain required to complete the path for the conduction and distribution of the impulse. The most important groups of the collaterals and stem-fibres of the posterior roots are: 1. The long ascending tracts passing chiefly to the nuclei of the medulla. 2. The fibres passing to the cells of the column of Clarke. 3. The collaterals passing to the anterior horn cells. 4. The fibres entering the posterior horn from the tract of Burdach and of Lissauer to end about the neurones of the II order situated within the gray matter of the posterior horn and the intermediate gray matter. The direct ascending posterior tract includes the dorsal root-fibres that pass uninterruptedly upward within the posterior column as far as the nuclei of the medulla. On entering the cord they lie at first within the tract of Burdach, but in their ascent are gradually displaced medianly and dorsally by the continued addition of other root-fibres from the succeeding higher nerves. In consequence, in cross Diagram showing e.\ directed inward. During their descent the fibres of the pyramidal tract give of! at diflerent levels col- laterals, which Ix-nd horizontally inward and forward, enter the gray matter, and end in rela- tion with the anterior horn cells. A similar course is followed by the parent fibres on reach- ing the segment for which they are destined, the terminal part <»f the individual fibres sweeping in short curves through the intervening ground- bundle of the lateral column to gain the radicular cells around which they end. By means of its collaterals, each pyramidal fibre establishes rela- tion with several cord-segments. The fibres of this tract are relatively tardy in acquiring their medullary coat, which process does not Ixrgin until the last month of fcetal life and is not com- pleted until after the second year. j| sixth thoracic Section of spin.n i" ; ,n segmt'"' • slender i>ostcrior coriiua covered with sui>- stantia Kclatinosa; postero-lateral angle marks greatest width of anterior comu. >' 6. Preparation by Pro- fessor ?piller. The direct cerebellar tract Tfas- ciculus ccrcbellospinalis;, is an important ascending jjath of the second order that establishes communication between the reception sensory cord-nucleus formed by Clarke's cells and the cerebellum. In cross-sections of the thoracic region, the tract forms a superficial flattened comet-shaped field that occupies the dorsal half of the lateral column, extending from the ape.x of the posterior horn forward along the periphery of the cord, to the outer side of the lateral pyramidal tract, to about the anterior plane of the gray commissure. Its ventral end, particularly in the lower cer\ical region, is broadest and projects somewhat into the lateral column in advance of the lateral pvramidal field. Although as a compact strand the direct cerebellar tract begins at the tenth thoracic segment, it is represented by isolated fibres in the lumbo- sacral region. The fibres collectively are large and become medullated about the sixth fcetal month (Bechterew). In a general way the fibres having the longest course occupy the dorsal part of the tract and those having the shortest the ventral ( Flatau ;. Arising as the axones of the cells of Clarke's column, the components of the tract pass in cur\-es almost horizontally outward through the gray matter and lateral column to the peripheral field, on gaining which they bend sharply brainward and ascend without interruption to the medulla. Their further course includes the pas- sage through the dorso-lateral field of the medulla as far as the inferior cerebellar peduncle, by which the fibres reach the cerebellum to end in relation with the superior worm, on, probably, both the same and the opposite sides. The tract of Gowers (fasciculus anterolatcralis superficialis) constitutes another pathway of the II order, which connects the cord with the cerebellum and probably also establishes relations with the cerebrum. In cross-sections the tract appears somewhat uncertainly defined owing to the»intermingling of its fibres with those of adjoining strands, but in the main it includes a superficial crescentic field that touches the direct cerebellar and lateral pyramidal tracts behind, extends along the margin of the cord for a variable distance, and usually ends in front in the vicinity of the ventral ner\'e-roots. The inner boundary, separating the tract in question from the lateral ground-bundle, lacks in sharpness and is overlaid by the adjoining strands. Below, the tract appears about the middle of the lumbar region and continues throughout the remainder of the cord. As Gowers' tract ascends, it fails to show the considerable increase in size that might be expected in view of the continual additions that it receives. In explanation of this, the probable mingling of some of its fibres with those of the direct cerebellar tract, rather than their ending in the cord, seems the most plausible (Ziehen ). The exact origin of the constituents of Gowers' tract is still uncertain, but it is ver^' likely that its fibres are chiefly the axones of the neurones (marginal and inner cells) situated within the posterior horn, partly from the same and partly from the w'lirri-: mattkr of thk simxal cord. 1045 Fig. S99 P^^- #;^' />j '■* opposite sides, with ct)nlril)utions, possil:)!)-, from the cells of the intermediate gray matter. After traversing the cord, the lateral field of the medulla, and the tegmental portion of the pons, the tract ascends the brain stem to the vicinity of the inferior cor- pora quadrigemina. Here the major part of the fibres turn backward and, by way of the superior cerebellar peduncle antl the superior medullarv \-elum, reach the cerebellum to entl mostly in the superior worm, jiartly on the same side and partly crossed ( Hochej. Possibly a part of the cerebellar contingent may share the path of the direct cerebellar tract and in this way reach the cerebellum by its inferior peduncle (Ziehen). It is probable that all fibres from Gowers' tract do not pass to the cerebellum, l)ut that some continue upward to terminate in relation with the neurones of the superior corpora quadri- gemina and of the optic thalamus. The fibres of the tract acquire the medullary coat about the beginning of the eighth month of foetal life (Bechterew). The lateral ground-bundle (fasciculus lateralis proprius) of Flechsig includes the remainder of the lateral column. Much uncertainty prevails as to its detailed paths, but beyond question the composition of the ground-bundle is very complex and comprises a number of long exogenous paths that descend from the brain, as well as one long ascending and many shorter endogenous strands, both ascending and descending. These short tracts occupy chiefly the central parts of the lateral column and, in a general way, lie close to the gray matter, within an area between the ante- rior and posterior horns, known as the boundary zone. They are, however, not limited to this field, as not a few of their fibres lie scattered among the longer exogenous tracts occupying the more lateral portions of the ground-bundle. --' - v^L*. Section of spinal cortl at level of lower part of fifth lumbar sejjment ; gray matter relatively large in amount ; anterior coriiua bulky. Preparation by Professor SpiUer. X 6. Fig. 900. One long endogenous path, the spino-thalamic tract, is of unusual importance since it estab- lishes a direct sensory link between the cord and higher centres. This tract arises from the cells of the posterior horn of the opposite side, the axones crossing in the anterior commissure to pursue a course brainward within the antero-lateral ground-bundle. Although the fibres of this tract are scattered and not collected into a compact strand, their chief location is just medial to Gowers' tract. Associated with the fibres destined for the optic thalamus are others {tradtis spino-tectalis) that end in the region of the corpora quadrigemina. The short endogenous tracts include both ascending and descending fibres which arise as the axones chiefly of the marginal and inner cells of the posterior horn, some coming from the opposite side by way of the posterior intracentral commissure. Entering the lateral column the axones undergo T-like division with ascending and descending limbs. The former pass upward for a distance that usually includes only from one to three segments, then bend inward and enter the gray matter to end probably in relation with other posterior horn cells. The down- wardly directed limbs form the descending endogenous fibres, which, in addition to occupying the boundar\- zone are also scattered among the longer tracts of the ground-bundle. After a relatively long course, they enter the gray matter to end probably in relation with the anterior horn cells. They are, therefore, regarded as establishing reflex-paths. Since these endogenous strands link together various levels of the cord, they are often collectively termed intersegmental association fibres. The exogenous tracts of the lateral ground-bundle are closely related with those found in the ground-bundle of the anterior column and what may be said of the former largely applies to the latter. Notwithstanding the study that these tracts have received, much uncertainty exists as to their exact origin and termination ; it may be stated in a general way, however, that they bring the higher sensory and coordinating centres into relation with the spinal cord and constitute, therefore, descending paths other than the Seciton of spinal cord at level of third sacral ses:ment ; posterior cornua with substantia gelatinosn are relatively bulky. Preparation by Professor Spiller. X 8. I046 liLMAN ANATOMY. pyramidal tracts. AmoiiK those wliose existence within the antero-lateral ground-lnindle may be consitiered as established, or at least probable, are the following: 1. Rubrospinal Jibrcs from the cells of the red nucleus within the cerebral peduncles. 2. Tectospinal Jibris from the cells of the anterior corpora (|uadrigemina. 3. Vestibulospinal fibres from the cells of the lateral vestibular (Deiters') nucleus. 4. Medullospinal fibres from the cells of the formatio reticularis, arcuate and lateral nuclei. 5. Olivo-spinal fibres from the cells of the inferior olivary nucleus. Of these strands, those from the red nucleus, corpora (juadrigemina, and vestibular nucleus, descend chielly within the lateral ground-bundle, whilst those from the medulla are j^articularly within the anterior ground-l)undle. Although the latter includes the greater part of the de.scending cerebello-rubro-sjiinal fibres in the narrow perii)heral sulco-marginal zone of Marie, other fibres are probably distributed within the lateral column in front of the direct pyramidal tract. These descending indirect cerebellar fibres are often collectively known as the tract of Marchi-Lowenthal. For the most part the exogenous strands are so intermingled and scattered that they are without definite outlines; an exception to this is presented by the olivary fibres, which are sometimes seen as a fairly distinct triangular bundle, Scciioii <>i spinal cord at level of J"^^ behind the anterior root-fibres at the periphery of the cord, fifiti sacral segment ; anterior cornua known as Helweg's tract. Concerning the exact ending of Uon'by^'professorspXr' :^l^^''''" tl^ese descending paths little is known, but it is reasonable to assume that they terminate at various levels in relation with the ventral horn cells which are thus brought under the coordinating influence of the higher centres. The Fibre-Tracts of the Anterior Column. — According to the simplest classification the anterior column includes two subdivisions : ( i ) the anterior pyra- midal tract and (2) the anterior ground-bundle. The anterior pyramidal tract (fasciculus cerebrospinalis anterior), also called the uncrossed ox direct pyramidal tract, stands in complcmcntal relation with the lat- eral pyramidal fasciculus, being composed of the pyramidal fibres that do not undergo decussation in the medulla oblongata. It usually contains about 15 per cent, of the pyramidal fibres, but may include a much larger proportion ; on the other hand, it may be entirely suppressed when, as rarely happens, total crossing occurs. The direct pyramidal tract occupies the inner part of the anterior column, forming a narrow area along the median fissure that extends from the Mhite commis- sure behind to near the ventral margin of the cord. Ordinarily the tract ends below about the middle of the thoracic cord, but in exceptional cases, when a larger pro- portion of the pyramidal fibres than usual is included in the tract, it may extend as far as the middle of the lumbar enlargement, with corres- ponding increase in its cross area. If, on the other hand, Fig. 902. the number of uncrossed fibres is unusually small, the tract may reach only as far as the cei"vical enlargement, with a reduction of its sagittal dimension. Although often spoken of as the ' ' uncrossed ' ' pyramidal tract, this characteristic applies only to the relation of the fibres at the decussation in the medulla, since in their downward journey in the cord ,evefo^\mVpart'ofcocc>4eli the great majority of the fibres traverse the anterior white segment; differentiation of cor- ■ • ' '.I 1. J- u'i' "ua is uncertain. Preparation commissure at appropriate levels to end in arborizations by Profe.ssor Spiiier. ^ s. about the ventral root-cells of the anterior horn of the opposite side. It is highly probable, however, that some fibres do not tindergo decussation, but terminate about tlie radicular cells of the same side. The anterior ground-bundle (fasciculus anterior proprius), following the divi- sion of Flechsig, includes the remainder of the ventral column. In front, where its lateral limits are uncertain, it is continuous with the ground-bundle of the lateral col- umn, the two together being often with advantage regarded as constituting a single antero-lateral tract. What has been said concerning the constitution of the lateral ground-bundle applies in the main to that of the anterior column, since, here as there, the region bordering the gray matter contains chiefly the short endogenous strands, while the more peripheral parts of the ground-bundle are occupied by the long exogenous paths, intermingled, however, with the longer intrinsic fibres. WHITE MATTER OF THE SPL\AL CURU. 1047 The endogenous fibres arise as the axones, chielly of the inner ceils uf the posterior horn, as well as from the cells of the intermediate gray matter (Ziehen), and in great measure cross by way of the anterior white commissure to the opposite anterior column. After undergoing T-division, their upwardly directed limbs cx)nstitute the ascending paths and those coursing downward the descending ones. While both sets of fibres for the most part pursue only a short patli, that of the de.scending limbs is usually the longer, the libres entering the gray matter t(j end in relation with the anterior liorn cells of lower levels. They are, therefore, regarded as .secondary reflex paths. The termination of the ascending limbs is uncertain, but probably is within the gray matter of the posterior horn. The exogenous tracts of the anterior ground-bundle, have been mentioned in connection with those of the lateral column. The investigations of Lciwenthal, Marchi, Bechterew, Thomas and others, .support the presence within the anterior ground-bundle, also within the lateral column, of long efferent (cortifugal) paths that arise, at least indirectly, from neurones within the cerebellum, and end in relation to the anterior horn cells. These paths, collectively known as the descending cerebello-spinal tract, or tract of Marchi- L'diventhal, are of uncertain extent and outline, and more or less mingled with the constituents of other strands. In a general way the descending cerebello-spinal fibres occupy a narrow crescentic field that appropriates the periphery of the cord for a variable distance both mesially and laterally. In the anterior column the tract includes the anterior marginal bundle, probably from the nucleus fastigii of the cerebellum of the same and opposite side, and mesially mingles with fibres from the corpora quadrigemina as con.stituents of the visual reflex paths. The termination of these descending paths is assumed to be in relation with the anterior horn cells, which in this manner are brought under the influence of the higher coordinating and reflex centres. In recapitulation the chief fibre-tracts of the spinal cord may be grouped as follows: I. Within the Posterior Column — Ascending Paths : Direct ascending posterior root-fibres. Ascending endogenous fibres. Descending Paths : Descending posterior root-fibres. Descending endogenous fibres. II. Within the Lateral Column — Ascending Paths : Direct cerebellar tract. Gowers' tract. Spino-thalamic tract. Short endogenous fibres. Descending Paths : Lateral pyramidal tract. Indefinite exogenous tracts (including the rubro-spinal, quadri- gemino-spinal, vestibulo-spinal, cerebello-spinal and olivo- spinal). Descending endogenous fibres. III. Within the Anterior Column— Ascending Paths : Ascending endogenous fibres from posterior horn cells. Ascending endogenous fibres from anterior horn cells. Descending Paths : Direct pyramidal tract. Descending cerebello-spinal fibres. Tegmento-spinal fibres. Blood-Vessels of the Spinal Cord. — The arteries supplying the cord are from many sources — the vertebral, deep cervical, intercostal, lumbar, ilio-lumbar and lateral sacral of the two sides — since the vascular net-work within the pia accoinpanies the nervous cylinder throughout its length. Above and within the skull, the verte- bral arteries give off the two anterior and the two posterior spinal arteries, of which the latter retain their independence and descend upon the dorso-lateral surface of the cord, one on eacli side, in front of the posterior nerve-roots. The two anterior spinal arteries, on the other hand, soon unite (somewhere above the level of the third cervical nerve) into a single trunk, which descends along the ventral surface of the cord, just in front of the anterior median fissure. I04S II L.MAN AN ATOM W Fig. 903. Posterior sulcal rostero-lateral Penetrating artery As these stems pass downward, they are joined and reinforced by tin- siffiiniila/ spinal branches j^iven off by the \ertel)ral, intercostal, lumbar and literal sacral arteries, which enter the spinal canal throuj^h the inter\ertebral foramina and, after piercinj^' the dura and L;i\in}^ ofT small radicular branches to the ner\e-roots them- selves, divide into \entral antl dorsal branches that follow the rcsj)ecti\e nerve-roots to the cord, where they join with the longitudinal trunks which the)- thus assist in maintaininj^. By the junction of horizontal branches arising from these arteries, a series of complete annular anastomoses is formed around the cord, which is still further enclosed by additional vertical stems resulting from the union of upward and downward coursing twigs. In this manner, in addition to the large single anterior spinal trunk {tractus arteriosus spinalis anterior') in the mid-line in front and the paired postero-Iateral trunk [tractns arteriosus postero-lateralis spinalis) just in advance of the dorsal nerve-roots, smaller longitudinal arteries are formed at the side and in the vicinity of the nerve-roots. From the arterial net-work within the pia, the nervous tissue is supplied hy pene- trating- t'iK'ios that enter the surface of the cord at \-arious ])oints. The gray matter receives its principal blood-suj^ply from the series of anterior Jissural arteries, over two hundred in number, which pass from the anterior spinal trunk backward within the median fissure to its bottom and there divide into right and left branches, which traverse the anterior white commissure to gain the gray matter on either side of the central canal. These vessels, the suleo-viar- ginal arteries, divide into ascending and descending branches that pro\ide for the entire gray matter with the exception of the most peripheral zone. The latter, together with the white matter, receives its supply from the pe7ietrating branches that come from the surrounding intrapial trunks and enter the surface of the cord. Unpaired horizontal twigs, the pos- terior sulcal arteries, follow the posterior median septum at different levels for some distance, but before reaching the posterior commissure usually break up into terminal ramifi- cations, some of which pass to the gray matter of the posterior horns. Communications e.xist between the penetrating twigs of the radicular arteries and the lateral branches of the anterior fissural. After entering the nervous tissue, however, each artery provides the sole supply for some definite jxart of the cord ; they are therefore "end-arteries," a fact which explains the extensive and elaborate system of Aessels necessary to maintain the nutrition of the cord. The plexiform veins within the spinal pia are formed by the union of the small radicles that collect the blood from the intraspinal capillaries and, after an independ- ent course similar to that of the arteries but not accompanying them, emerge at the surface of the cord. From the venous net-work within the ])ia six main longitudinal trunks are differentiated. These are : — the unpaired anterior median vei?i, in front of the corresponding fissure ; the paired antero-lateral veins, just behind the ventral nerve-roots — these two sets receiving the tributaries emerging from the median fissure and in the vicinity of the anterior root-fibres ; the unpaired posterior median vein, behind in the mid-line ; and the paired postero-latcral veins, just behind the dorsal roots. The blood is conveyed from these intrapial channels chiefly by the radicular veins, following the nerve-roots, which communicate with or terminate in the anterior and posterior longitudinal spinal veins within the vertebral canal, from which the Anterior fissural Anterior Ascending branch spinal artery Part of transverse section of injected spinal cord showiiij; vascular sujiply of white and gray matter. • 10. W'lliri-: MATTER OF THE SPINAL CORD. 1049 intervertebral efferents carry the blood into the vertebral, intercostal, lumbar and lateral sacral veins. A part of the blood from the intrapial plexus is conducted upward by the anterior and posterior median veins into the venous net-work covering the pons and thence into the lower dural sinu.ses. Definite lymphatic vessels within the spinal cord are unknown. Development of the Spinal Cord. — A .sketch of tlie j^eneral hislo^enetic |)rocesses leading to the differentiation of the neurones and tlie neuroglia has l^een given (page 1009) ; it remains, tlierefore, to consider liere the changes in the neural tube by which the defuiite spinal cord is evolved. From the time of its closure, probably about the end of the second week of fietal life, the neural tube presents three regions : — the relatively thick lateral ivalls and the thin Neu- tral and dorsal intervening bridges, \\\ki floor- and roof-plates, that in front and behind complete the boundaries of the canal in tlie mid-line. By the fifth week the lateral walls e.xhibit a distinct differentiation into three zones — the inner epeudymal layer, the middle nuclear layer and the outer uiargiual layer, surrounded by the external limiting nievibrane. In contrast to the other two, the marginal zone is almost devoid of nuclei and, beyond affording sup])ort and perhaps assisting in providing a medullary coat, plays a passive role in the production of the nervcnis elements. By this time the former general oval contour of the developing cord, as seen in cross-sec- tions, has become nK)dified by the conspicuous thickening of the antero-lateral area of the nuclear layer into a prominent mass on each side, whereby the reticular marginal layer is pushed out- FiG. 904. Roof-plate Fig. 905. Roof-plate Dorsal zone Dorsal root-fibres Neuroblasts ■ Floor-plate Floor-plate Ventral root-fibres Developing spinal cord of about four weeks. X 100. (His.) Developing spinal cord of about five weeks. X 60. (His.) ward with corresponding increase in the width of the entire ventral part of the cord, which is now broadest in front. Within this thickened ventro-lateral part of the nuclear layer, later the anterior horn of gray matter, as early as the fourth week young neurones are seen from which axones grow outward through the marginal zone and pierce the e.xtemal limiting membrane as the representatives of the anterior root-fibres of the spinal nerves. Postero-laterally the thin nuclear layer is covered by a somewhat projecting thickened area within the marginal layer, known as the oz'al bundle, whose presence is due to the ingrowth of the developing dorsal root- fibres from the sensory neurones of the spinal ganglion, which process begins as early as the end of the fourth week (His). i\ssociated with these changes, the lumen of the cord becomes heart-shaped in consequence of a conspicuous local increase in its transverse diameter, with corresponding bulging of the lateral wall. In this manner a longitudinal furrow appears by which the side walls of the tube are differentiated into two tracts, the dorsal and the ventral zones (the alar and basal laminae of His). This subdivision is of much importance, since in the cord-segment, and also with less certainty in the brain-seginent of the neural tube, these tracts are definitely connected with the root-fibres of the spinal nerves, the dorsal zone with the senson,- and the ventral zone with the motor roots. In advance of the floor-plate the ventrally protruding halves of the cord include a broad and shallow furrow which marks the position of the anterior median fissure. During the sixth week the form of the tube-lumen becomes further modified by the elongation and narrow- I050 HIM AX ANATOMY. ing t)f the dorsal part of the canal in conseciuencc of tliL- approximation of its walls, which in the course of the seventh week is closer and, by the end of tiie second montii is < ompleled by the meeting and fusion of the adjacent inner layers, with obliteration of the intervening cleft and the jiroduction of the posterior median septtnn in its place. Since tlie partition is formed by the union of the inner (ependymal) layers, it is probable that the septum is to be regarded as essentially neurogliar in origin and character, it must be remembered, lunvever, that a certain amount of mesoblastic tissue may be later iiUrotluced in company with the blood-vessels which subsequently invade the septum. The remaining and unclosed part of the lumen for a time resemi)les in outline the conventional spade of the playing card, with the stem directed ventrally ; but later gradually diminishes in size and accjuires the contour of the definite central canal. During these alterations in the e.xtent and form of its lumen, the i!;ray matter of the develop- ing cord markedly increases, esjiecially behind where the posterior horn appears as a projection beneath the broadening mass of the ingrowing dorsal root-fibres. As the posterior horti becomes better defineti, the root-bundle becomes meso-laterally displaced, lying behind the horn, and then constitutes the traet of Biirdach. Goll's tract is formed somewhat later and at about the third month appears as a narrow wedge-shaped area tiiat is introduced between the mid-line and Hurdach's tract. Towards the end of the second month, the anterior luhite commissure is indicated by the oblique transverse ingrowth of a.xones into the most ventral part of the floor- plate as they make their way to the opposite side. Meanwhile the anterior median fissure has Fig. 906. . Posterior column Fig. 907. Ventr.il zone Goll's tract Burdach's tract I'osterior median se|>tum - Root-fibres Anterior colunm Developing spinal cord of about seven and one-half weeks. X 44- {His.) Anterior median fissure with pial process Developing spinal cord of about three months. X 30. ( Hts. ) become deeper and narrower in consequence of the increased bulk of medio-VL-ntral parts of the cord. As tile fissure is thus ditYerentialed the process of mesoblastic tissue, which from the earliest suggestion of the groove occupies the depression, is correspondingly elongated and affords a passage for the blood-vessels destined for the nutrition of the interior of the cord. Until the third month the gray matter, derived from the nuclear layer, is much more voluminous than the surrounding marginal layer, which, so far as the contribution of nervous elements is concerned, is passive, since its conversion into the white matter depends upon the ingrowth of axones from the neurones situated either within or outside the cord. The development of the individual fibre-tracts includes two stages, between the comple- tion of which a considerable, and .sometimes a long, period intervenes. The first marks the invasion of the supporting tissue of the marginal zone by the ingrowing axones as naked axis- cylinders ; the second witnesses the clothing of these fibres with myelin. The period between the appearance of the tract and the development of the medullary coat is variable. In .some cases, as in the great cerebro-spinal motor paths, although the fibres grow into the cord during, the fifth month of fcetal life, myelination does not begin until shortly before birth and is not completed until after the second year. In other cases, as in the direct cerebellar, a period of three months, from the third to the sixth, elapses. It is probable that the acquisition of the medullary coat commences before the functional activity of the fibres begins, although such stimulation undoubtedly assists ; further myelination proceeds gradually along the course of the fibres and in the direction of conduction. PRACTICAL CONSIDRRATIONS: SPINAL CORD. 1051 Based on the ohservaliuiis oi Flcchsij;, His, Hcchtercw, and olliers, tlie time of the api)earance and of the tlcvelopment of tlie medullary coat of some of the fibres within the spinal cord may be given. Fibres of Appear Myelinate Anterior root about 4th week durinj^ 5th montii Burdach's tract durinj^ 4th week end of 6tli month (ioll's tract about 9th week beginning;; of 71I1 month Pyramiilal tracts end of 5lh month 91!) month to 2nd year Direct cerebellar tract bei^inninj;- of 3rd montii aljoiit 6th month Cowers' tract durini; .jth iiionlli durinjr 61I1 iiionlli The presence of the si/i/is tennina/is (page 1030) in the cord at birth depends partly wyton the persistence of the lumen of the central canal at the lower end of the conus meclullaris and partly upon a i>ioiiferation of the wall-cells of the subjacent segment, followed by secondary dilatation shortly before birth. During the early weeks of deveioi)ment, the neural tube extends to the lowermost limits of the series of somites ; but after differentiation of the root-fibres begins, the segment of the cord below the level of origin of the first coccygeal nerves is marked by feeble proliferation, the effects of which are soon manifest in the rudimentary condition of the caudal end of the cord. With the subsequent development of the other regions, this iiistological contrast becomes more e\ident, to which is soon added the consj^icuous attenuation caused by the attachment of the lower end of the cord to the caudal pole of the spine, which elongates with greater rapidity than the contained nervous cylinder. In this manner the lowest segment of the cord, with its mesoblastic envelope, is converted into the delicate thread-like fi/um ienninale, within whose upper half are found the remains of the rudimentary nervous tissue. PRACTICAL CONSIDERATIONS : SPINAL CORD. Congenital Errors in Development. — The spinal cord may be absent {a?fiyelia), or it may be defective in a certain portion {ateomyelia). In such conditions, however, the patient cannot Hve. The cord may be double from bifurcation {dipiomye/ia). A spina bifida is a congenital condition due to a deficiency in the vertebrae, almost always of the laminae and spinous processes. There is usually a protrusion of the contents of the spinal canal, although in some cases there is no protrusion, and in others the vertebral canal, or even the central canal of the cord may be open to the surface. Three varieties of tumors are described according to their contents. If the meninges only protrude from the canal in the form of a sac containing cerebro- spinal fluid, it is called a meningocele ; if the sac contains a portion of the cord also it is called a meningo-myelocele. In the third variety, syringo-myelocele, the cavity of the tumor is found to consist either of the dilated canal of the cord, so that the thinned-out substance of the .cord is in the wall of the sac, or of a cavity in the cord tissue itself. This is the least common of the three forms. In the nieningo-7nyelocele , which is the most common form, the cord becomes flattened out and attached to the posterior wall of the sac, but still has its central canal intact. The spinal nerves cross the sac to their corresponding intervertebral foramina. In this and in the syringo-myelocele there is frequently some degree of paralysis in the parts below from disturbance of the cord at the seat of the tumor. The most common seat of the defect is in the lumbo-sacral region. It is rare in other parts of the spine. Therefore, the bowels, bladder, and lower extremities are the parts most frequently affected. If the lesion is confined to' the lower part of the sacral region, the extremities usually escape. Paralytic talipes is comparatively common. There is no sharp line of demarcation between the medulla oblongata and the cord. The beginning of the latter is variously given as at the origin of the first cer\'ical nerve, the lower margin of the foramen magnum, or the decussation of the pyramids, the last being the more generally accepted. Since in the adult, the spinal cord ends below usually at the level of the disc between the first and second lumbar vertebrae, injuries of the spine below the second lumbar vertebra do not involve the cord. The membranes of the cord, however, containing cerebro-spinal fluid extend as far as the second or third sacral vertebra, so that at this level injuries with infection may cause fatal meningitis. I052 HIMAX ANATOMY. The bcny canal is lined with periosteum, unlike the cranium, in which the external layer of the dura mater serves that purpose. The spinal dura is separated from the posterior common lij^ament, the ligamenta subfla\a, and the ixriosteum by a fatty areolar tissue containing a plexus of veins. Extensive extradural hemorrhage may, therefore, occur without serious [pressure on the cord. The blood tends to sink by gravity, and later may jjroduce symptoms of compression. The dura is thick and strong and offers considerable resistance to the invasion of disease from with- out, even to tuberculosis with caries of the vertebrce, or to malignant tumors arising within the vertebrae. Infections outside the spinal column, as in abscess of the back, or bed sores, may extend along the communicating veins, giving rise to extradural abscess and perhaps to extensive meningitis. The spinal cord, surrounded by cerebro-spinal fluid, hangs loosely within the dura, being attached to it only by the roots of the spinal nerves which receive invest- ments from the dura as they pass outward, by the ligamenta denticulata, and by the delicate fibres of arachnoid tissue extending from the pia to the dura. The cord is, therefore, not frequently injured from external violence. The numerous articulations of the vertebrse and the elasticity of the ligaments and of the intervertebral discs permit the distribution of much of the force applied to the spine before it reaches the cord. The greater part of the cerebro-spinal fluid is contained in the subarachnoid space, which communicates freely with the same space in the cranium, and is con- tinuous with the ventricular fluid through the foramen of Majendie. The cord is exposed to the danger of penetration by sharp instruments only from behind, but even here the overlapping of the lamiuce and spinous processes offers an excellent protection. This protection is largely lacking above and below the atlas, and the risk there from such wounds is correspondingly greater. At lower levels in order that the canal may be reached, the vulnerating instrument must be directed in the line of the obliquity of the laminae, which will var}- in the different portions of the spine, being greatest in the dorsal region. Co)iaissio7i — shaking with molecular disturbance and without obvious gross lesion — fjf the cord, although more frequent than has been supposed, is rare because of {a) the arrangement of the different constituents of the vertebral column, which by means of its cunes, the elastic intervertebral discs, its numerous joints, and the large amount of cancellous tissue in the vertebral bodies, is able to take up and distribute harmlessly forces of some degree of violence ; {b) the situation of the cord in the centre of the column, where, as the most frequent serious injuries to the spine are caused by extreme forward flexion, it is somewhat removed from danger in accordance with a law of mechanics that " when a beam, as of timber, is exposed to breakage and the force does not exceed the limits of the strength of the material, one division resists compression, another laceration of the particles, while the third, between the two, is in a negative condition" (Jacobson ) ; ( c) the suspension of the cord in the surrounding cerebro-spinal fluid ("like a caterpillar hung by a thread in a phial of water" — Treves) by its thecal attachments and nerve-roots ; (d) its connection above with the cerebellum, itself resting on an elastic "water-bed" which minimizes the transmission downward of violence applied to the cranium. Many of the cases reported as concussion are undoubtedly due to hemorrhage or other gross lesions of the cord. Conhisioyi of the cord may occur from sprains, as in forced flexion of the spine. The most frequent and most serious cases are those due to fracture-dislocations of the spine, the cord being more or less crushed between the upper and lower fragments. It is so delicate a structure that it may be thoroughly disorganized without evident injur\' to the membranes or alteration of its internal form. The paralysis of the parts below will be complete or partial according to whether the whole or only a part of the transverse section of the cord at the seat of injury is destroyed. Since when the lesion is complete everything supplied by the cord below the seat of the lesion is paralyzed, the higher the injurv' to the cord the greater the gravity of the case. When the atlas or axis is fractured and displaced the vital centres in the medulla are in danger and death may result immediately. The phrenic nenes which arise chiefly from the fourth cer\-ical segment, but partly from the third and fifth segments, are also paralyzed and respiration ceases. PRACTICAL COXSIDKRATIONS : SPINAL CORD. 1053 \x\ fracture-dislocations of the spine it is the body of the vertebra which is most frequently fractured, the liijainents yieltlinjj^ posteriorly and permitting the dislcjcation. The fractured edi^es of bone are, therefore, in front of the cord ; and, as the upper fragment passes forward, the anterior or motor portion of the cord is pressed and crushed against the sharp upper edge of the lower fragment. In partial transverse lesions of the cord the paralysis below the lesions affects, therefore, the motor columns of the cord more than the sensory columns which are in part posterior. The most frequent seat of fracture-dislocation of the spine is in the thoraco- lumbar region (page 145). Fortunately, it is this variety which offers the best prognosis, since the cord ends usually just below the lower border of the first lumbar vertebra, and the cauda equina being more movable and tougher than the cord itself, it can better evade the encroachment on the canal, although in spite of these facts, it is not infrequently injured in such lesions. The bodies of the lumbar vertebrie are the largest and most cancellous, the intervertebral discs the thickest and most elastic, so that crushing of them occurs with less tendency to invade the canal and injure the cord than in any other portion of the sj)ine. In caries of the spine (Pott's disease) the lesion is situated in the bodies of the vertebne, and therefore, in front of the cord. As the inflammatory exudate extends it will invade the spinal canal anteriorly, often producing an external pachymeningitis. The irritation and pressure resulting will again aftect the motor portion of the cord, first producing a paralysis of motion in the parts below, varying in degree according to the amount of pressure on the cord. If sensation is impaired it is a later phenomenon and is due to greater pressure upon the cord, and in some cases to myelitis. The loss of motion is often the only effect produced. If the lower cervical region is involved by the lesion the phrenic nerves will escape paralysis, but the arms, trunk, bladder, rectum, and lower extremities will be affected. Since the intercostal and abdominal muscles are involved in the paralysis, breathing will be difficult and will depend upon the action of the diaphragm only. Thus as the lesion occurs at successively lower levels, the highest limits of the paralyzed area descend, and the expectation of life increases. In the cervical and thoraco-lumbar regions where the injuries to the spine and the cord are most frequent, are situated the two enlargements of the cord. The cervical begins at the fourth cervical vertebra, gradually reaches its largest diameter opposite the fifth and sixth vertebrcC, and then gradually decreases to the first thoracic, where it merges into the thoracic portion of the cord. Only in the thoracic region does the circumference of the cord remain the same throughout. The lumbar enlargement is shorter than the cervical and begins opposite the tenth thoracic vertebra, gradually increases to the twelfth thoracic, after which it gradually decreases to the conus meduUaris. The localization of lesions of the cord, producing symptoms of paralysis, will depend upon the height and extent of the paralyzed areas. It must be borne in mind that the nerve-roots arise from the cord usually at a level higher than the foramina through which they escape from the spinal canal. The first and second cer\'ical nerve-roots pass out of the canal almost horizontally. The intraspinal course of the succeeding nerve-roots increases gradually in obliquity so that the spinous processes of the second, third and fourth vertebrae correspond approximately to the level of the third, fourth and fifth cervical nerve-roots. The seventh cervical spine corresponds to the first thoracic ner\'e-root. The spinous process of the fifth thoracic vertebra is on a level with the seventh thoracic nerve, and the spine of the tenth thoracic vertebra with the origin of the second lumbar nerve. The first lumbar nerve arises just below the ninth thoracic spine, the second lumbar nerve opposite the tenth thoracic spine, the third and fourth lumbar nerves opposite the eleventh spine, and the fifth lumbar and the first sacral ner\es between the eleventh and twelfth thoracic spines. Only the spinous processes can be our surface guides, and it must be borne in mind that they are not always on the level of their corresponding vertebrae. Briefly, it may be said that the eight cer\^lcal ner\'es arise from the cord between the lower margin of the foramen magnum and the sixth cervical spine, the first six thoracic I054 HTMAN ANATOMY Fic. First cer\'ica1. vertebra -^ -Skull First thoracic _ vertebra ' • First thoracic spine V nerves between the hitter spine and the foiirtli thoracic, the lower six thoracic nerves between the fourth and ninth dorsal spines, the five lumbar nerves opjiosite the ninth, tentii and elexenth sj)ines, and the li\e sacral nerves opposite the twelfth thoracic and the first lumbar spine. A convenient rule to locate the levels of origin of the nerve-roots, applicable to the prelumbar nerves, is given by Ziehen as follows : — P'or the cervical nerves, subtract one from the number of the nerve, the remainder indicating the correspond- ing spinous process ; for the upper ( I-V ) thoracic ner\es subtract two ; for the lower (VI-XII) thoracic nerves subtract three. All the cer- vical nerves pass out through the intervertebral foramina above the vertebrae after which they are named, except the eighth cervical, which emerges between the seventh cer- vical and the first dorsal vertebrae. All the other spinal nerves escape below the vertebrae from which they are named. Since the nerve-roots pass a considerable distance down- ward within the spinal canal before leaving it, it follows that a lesion of the cord at a gixen le\'el, as from a fracture-dislocation of the spine, may be associated with a paralysis of the nerve-roots passing out at or below that level, and arising from the cord at a higher point. This must be taken into account in determining the seat of the lesion, since when the ner\'e-roots are not involved the lesion will be as much higher than its corresponding inter- vertebral foramina ( as indicated by the upper limits of the paralyzed area) as the length of the intraspinal course of the corresponding nerve- roots. Each root-cell in the anterior horn of gray matter is connected with a motor fibre, which passes out in the anterior root of a spinal nerve to its muscle. Motor impulses originating in the cortex of the brain, pass downward along the antero- lateral columns of the cord, chiefly in the lateral pyramidal tract. They first traverse the ganglion cells of the anterior horns before passing out in the anterior or motor roots to their destination. These ganglion cells constitute, at least functionally, the trophic centres for the muscles. Lesions of the anterior horns, therefore, besides causing First lumbar vertebra / First lumbar spine W First sacral vertebra Coccyx < DiaRram, based on frozen section, showing relations of bodies and spines of vertebrje to levels at which spinal nerves escape from vertebral canal. THE BRAIN. 1055 paralysis {polio-myelitis), will lead to atrophy of the corresponding muscles. The vasomotor centres are also in the anterior horns, probably in the intermedio-lateral tract. Sensory impulses pass to the posterior horns through the posterior roots, and some of them soon cross to the opposite side of the cord, others ascending in the posterior column. The lemniscus is probably the chief sensory tract in the medulla oblongata, pons, and cerebral j)eduncles. Kvery servient of the spinal cord contains centres for certain groups of muscles, and for reflex movements associated with them. A reflex begins in the stimulation of a sensory nerve. The impulse thus created passes to a centre in the cord and thence is transmitted to a motor nerve, thus producing a contraction of the muscle supplied by that nerve. The complete path of this impulse is called a reflex arc. The sensory impulse may be transmitted to different segments of the cord and thence out through the corresponding motor roots. Thus a complicated reflex arc is produced. It is to be assumed, however, that the impulse will take the shortest route, so that simple reflexes will have their reflex arc chiefly in those segments of the cord in which the posterior root enters. Each segment of the cord is connected with fibres from the brain to which must be ascribed the function of reflex inhibition. If the inhibitory fibres are irritated, the reflexes are impaired from stimulation of inhibition. If the conductivity of these fibres is destroyed, the reflexes are increased; but if the reflex arc is broken at any point, the reflexes are lost. Among the most important of these are the skin and tendon reflexes. The centres for the bladder, rectum, and sexual apparatus, are located in the sacral segment of the spinal cord at and below the third sacral segment. They regulate the functions of these organs and are associated in some unknown way with the brain. (See mechanics of urination, page 19 14). Htrmato-rliachis, or hemorrhage into the membranes of the cord fextramedullary hemorrhage), may result from an injury to the spinal column, as a fracture or a severe sprain. The bleeding may be from the plexus of veins between the dura and bony wall of the canal (most frequent), or from the vessels between the dura and the cord. In either case the symptoms will be much the same. There will be a sudden and severe pain in the region of the spine, diffused some distance from the seat of the in- jury, due to irritation of the meninges, and pain transferred along the distribution of the sensory nerves coming from the affected segments of the cord, accompanied by abnormal sensations, as tingling and hyperaesthesia. In the motor distribution there will be muscular spasm, or sometimes a persistent contraction of the muscles. Gen- eral convulsive movements, retention of urine, and, later, symptoms of paralysis may appear, but as a rule the latter is not complete. Hcemato-myelia, or hemorrhage into the substance of the cord (intramedullary hemorrhage) from traumatism, usually occurs between the fourth cervical segment of the cord and the first dorsal (Thorburn), and is commonly due to forced flexion of the spine, which is most marked in this region, as in falls on the head and neck. The cord has been crushed in such accidents without fracture of the spine and with only temporary dislocation. The hemorrhage is usually chiefly in the gray matter and may be only punctate in size, or may be large enough to extend far into the white matter, or even outside the cord into the subarachnoid space. The symptoms usually appear immediately after the injury and are bilateral, suggesting a total transverse lesion. There will be much pain in the back, occasionally extending along the arms or around the thorax. Spasms, rigidity, and paralysis rapidly ensue, with loss of the reflexes in the segment of the cord involved. There may be the same dissociation of sensation as in syringomyelia when the hemorrhage is confined to the centre of the cord. THE BRAIN. The brain, or the encephalon, is the part of the cerebro-spinal axis that lies within the skull. It is produced by the differentiation of the cephalic segment of the neural tube. Although the brain is often of great relative bulk and high complexity, as in man and some other mammals, it must not be forgotten that the spinal cord is the 1056 HUMAN ANATOMY. fundamental antl essential part of the nerxous axis and that the dej^jree to which the brain is tieveloi)ed is, in a sense, accidental and dependent upon the necessities of the animal in relation to the exercise of the hi},dier nerv(jus functions. In the lowest \ertel)rates, the fishes, in which association of the imjjressions received from the outer world is only feebly exercised, those parts of the brain rentlerinj^^ such functions jjossible, as the cerebral hemisi)heres, are \ery imperfectly rejjresented. On the other hand, in man, in whom the ca])acity for the exercise of the hijL^her nervous functions involvin.n association is consijicuous, the antero-superior parts of the brain, the pallium, as the rei^ions jjarticularly concerned are called, are so enormously developed that the human brain is thereby distinijuished from all others. Whether of low or high develoj^ment, all brains are evolved from certain fundamental parts, the brain-vcsichs, differentiated in the head-end of the embryonic neural canal ; the underlviuii conception of the brain, therefore, is that of a tube, bent and modified to a variable deijree by the thickenintj:, uneciual growth and expansion of its walls. Even when most complex, as in man, the adult organ exhibits unmistakable evidences of subdi\ision correspontling more or less closely with the ])rimary brain-vesicles, and contains spaces, the \entricles, that represent the modified lumen of these segments. Fig. 909. Orbital surface of frontal lobe Optic commissure Optic tract Cerebral peduncle Interpeduncular space Medulla Cerebellum Olfactory tract Stalk of pituitary body Tuber cinereum Mammillary bodies Cerebral peduncle Temporal lobe Pons Cerebellum Occipital lobe Spinal cord Simplified drawing of brain as seen from below, showing: relations of brain-stem to spinal cord and cerebrum. Preparatory to entering upon a description of the fully formed brain, it is desirable to consider briefly the broad plan according to which the organ is laid down and the general lines along which its evolution proceeds. Before doing so, however, it will be necessary to take a general sur\'ey of the relations of the several divisions composing the brain. Denuded of its investing membranes and the attached cranial nerves, and viewed from below (Fig. 909), the encephalon is seen to consist of a median brain-stem, that inferiorly is directly continuous with the spinal cord through the foramen magnum and above divides into two diverging arms that disappear within the large overhang- ing mass of the cerebrum. The brain-stem includes three divisions, the inferior of which, the medulla oblons;ata, is the uninterrupted upward prolongation of the spinal cord and above is limited by the projecting lower border of the quadrilateral mass THK BRAIX. ^37 of the next division, the pons Varolii. Beyond the ujjper margin of the pons the brain-stem is represented by a third division that ventrally is sej^arated by a deep recess into two di\erv;in<4 Hmbs, the cerebral peduncles^ or crura cerebri, to corre- spond with the hakes or liemisplures of the cerebrum, each of which receives one of the crura and in this manner is connected with the lower levels of the cerebro- spinal axis. The i^reater part of the medulla and j>ons is covered dorsally by the cerebellum, whose larg^e lateral expansions, or hemispheres, project on either side as conspicuous masses, distinj^uished by the closely set plications and intervening^ fissures that mark their surface. Of the five component parts of the brain — medulla, pons, cerebral peduncles, cerebrum, and cerebellum — the last two are coated with the cortical gray matter, in which, broadly speaking, are situated the neurr^nes that constitute the end-stations for the sensory impulses con\eyed by the \arious corticipetal paths and the centres controlling the lower-lying nuclei of the motor nerves. The brain-stem, on the other hand, whilst containing numerous stations for the reception and distribution of sensory impulses, is primarily the great pathway by which the cerebrum and the cerebellum are connected with each other and with the spinal cord. Viewed in a mesial sagittal section (Fig. 910), each of these divisions is seen to be related to some part of the system of communicating spaces that, as the lateral and third ventricles, the aqueduct of Sylvius and the fourth ventricle, extend from the cerebral hemispheres above, through the brain-stem and beneath the cerebellum. to the central canal of the spinal cord below. Since the lateral ventricles are two in number, in correspondence with the cerebral hemispheres in which they lie, their position is lateral to the mid-plane and hence only one of the openings, xh^foramifia of Monro, by which they communicate with the unpaired and mesially placed third ventricle, is seen in sagittal sections. Both the roof and the floor of the irregular third ventricle are thin, whilst its lateral walls are formed by two robust masses, the optic thalami, the mesial surface Corpus callosum Septum lucidum Fig. 910. Frontal lobe, mesial surface Anterior commissure Foramen of Monro I^amina cinerea Optic commissure' Pitiiitar}- i noor of third ventricle Mammillar}- body Aqueduct of Sylrius Pons Fourth ventricle Optic thalamus, dorsal surface Lateral wall of third ventricle (optic thalamus) Cerebral peduncle Roof of Sylvian aqueduct '- Occipital lobe Superior medullary \ elum White core of cerebellum Inferior medullary velum Medulla • Spinal cord ■ Simplified drawing of brain as seen in mesial section, showing relation of brain-steni. cerebrum and cerebellum, and ventricular spaces. of one of which forms the background of the space when viewed in sagittal section. The roof (A the ventricle is \ery thin and consists of the delicate layer of epeyidyma, as the immediate lining of the \entricular spaces is designated, supported by the closely adherent fold of pia mater which in this situation pushes before it the neural wall and contains within its lateral border a thickened fringe of blood-vessels, the 67 I058 lUMAN ANATOMY. choroid plexus. The two structures, the eijeiidyma and the pia mater together, constitute the nieinl)ranous velum interposituni that fcjrms the roof of the ventricle and hes beneath tlie triangular A-'/vz/.r, w hose \aultetl form is suggested by the arching ridge that descends in front of the thalamus and marks the position of the anterior pillar of the fornix. Behind, ju.st over the upper KtwiX of the Sylvian afjueduct, lies the cone-shaped pineal body that belongs to the third ventricle, from wliich it is an outgrowth. '\\\(: Jloor of the ventricle is also, for the most part, relati\ely thin and irregular in contour. It corresponds to the median part of the lozenge-shaped area, the interpeduncular space, which, seen on the inferior surface of the brain, is bounded behind by the anteriorly di\erging cerebral peduncles and in front bv the optic chiasm and the posteriorly di\erging optic tracts. The posterior half of this area includes the deep triangular recess at the bottom of which is seen the numerous minute o])en- ings of \\\Q: posterior perforated space through which small branches of the posterior cerebral arteries pass to the ojjtic thalamus and the crura. Passing forward, the paired corpora mammillaria, the tuber ciiiereum, the stalk of the pituitary body occupy successively the interpeduncular space. Anteriorly, between the trans- versely cut optic chiasm below and the recurved portion of the great arching com- missure, the corpus callosum, abo\e, the third ventricle is closed by a thin sheet of nervous substance known as the lamina cinerca. Through the foramina of Monro the lateral ventricles open into the third, and the latter communicates with the fourth ventricle by way of the Sylvian aqueduct. This narrow canal is surrounded below and laterally by the dorsal part or tegmentum of the cerebral peduncles ; above it lies a plate of some thickness the dorsal surface of which is modelled into two pairs of rounded elevations, the superior and inferior corpoj'a quadrigcmina. In sagittal section, the fourth ventricle appears as a triangular space, the anterior or basal wall being formed by the dorsal surface of the pons and medulla and the posteriorly directed apex lying beneath the cerebellum. The upper half of the thin tent-like roof of the ventricle is formed by the superior medullary velum, a thin layer of white matter that stretches from beneath the inferior corpora quadrigcmina to the cerebellum. A similar lamina, the inferior medullary vehun extends from the cerebellum downward, but before reaching the dorsal surface of the medulla becomes so attenuated that this part of the \'entricular roof, known as the tela ehorioidca, consists practically of the pia mater, although the ependyma excludes the vascular membrane from actual entrance into the ventricle. The pia, however, pushes in the ependymal layer and in this manner produces the vascular fringes known as the choroid plexus of the fourth ventricle. When view-ed from behind, the ventricle exhibits a rhomboidal outline, the lateral boundaries above being formed by two arms, the S2(perior cerebellar peduncles, that di\'ergingly descend from the sides of the corpora cjuadrigemina to the cerebellum. Similar bands, the inferior cerebellar peduncles, convergingly descend from the cerebellar hemisj^jheres to the posterior columns of the medulla and form the lower lateral boundaries of the ft)urth ventricle. Seen from directly abo\e (Fig. 984), the cerebrum, di\ided into its hemi- spheres by the deep sagittal fissiirc, is the only part of the brain visible, the other four divisions being masked by the enormously developed overhanging cerebral mantle. The effects of this expansion in displacing base-ward parts which, temporarily in man and permanently in the lower vertebrates, occupy a superior position, are conspicuous when the sagittal section of the developing (Fig. 913) and that of the fully formed human brain (Fig. 910) are compared. It should be noted, that although in the latter the brain-stem and the cerebellum are completely overhung by the cerebral hemispheres, they still are in relation with the free surface of the brain, and by passing beneath the posterior part of the cerebrum the dorsal surface of the cerebellum and of the brain-stem may be reached without mutilation of the ner\ous tissue. THE GENERAL DEVELOPMENT OF THE BRAIN. Even before complete closure of the anterior end of the neural tube, which takes place probably shortly after the end of the second week of foetal life, the cephalic region of this tube, slightly flattened from side to side, exhibits the results GEXKRAL I)i:\i:L()rMi:.\T Ol' Tin: I'.RAIX. 1059 of iiiKHjual i^rowtli ill two >liolu ajiislricliuns scparatiny ihixc dikilalions known as the primary brain-vesicles. The posterior of these, the hind-brain,' is much the Ioniser, exceedino the cxjnibined lenj^th of the other two (Fi}^'. 911;; altera short time when viewed from l)ehitul it presents an elongated lozen^^e-shaped form and, hence, is also calKtl the rhombencephalon. ihe middle vesicle, the mid-brain^ or mesencephalon, is cons]ncuous on account of its rounded form and pnnninent position, lying, as it does, over the marked primary flexure which the head-end of the neural tube very early exhibits. The anterior vesicle, known as the fore-brain, or prosencephalon, at lirst is small and rounded, but soon becomes modified by the appearance, on either side, of a hollow protuberance, the optic vesicle, that pushes out from the lower lateral wall. Por a time the optic vesicle communicates with the main ca\ity of the fore- brain by a wide opening. This gradually becomes reduced and constricted until the Fore-brain allium Mid-brain Optic vesicle Fig. 911. Fore-brain (thalamic region > Pallium Mid-brain Hind-brain Reconstruction of b'-ain of human embryo of about two weeks (3.2 mm.); ./4, outer surface; B^ inner surface; np, neural pore, where fore-brain is still open ; cs, anlage of corpus striatum ; o>\ optic recess leading into optic vesicle; A/, hypothalamic region. {His.) evagination is attached by a hollow stem, the optic stalk, which later takes part in the formation of the optic ner\e that connects the eye with the brain, the vesicle itself giving rise (page 1482) to the nervous coat of the eye, the retina. By the time the optic evagination is formed, the front part of the fore-brain shows a slight bulging, narrow below and broader and rounded above, and separated from the optic outgrow^th by a slight furrow. This is the first suggestion of the anlage of the hemisphere or pallium (His). The latter soon gives rise to two rounded hollow protrusions, one on either side of the fore-brain, that rapidly exixmd into the conspicuous primary cerebral hemispheres. The lower part of the fore-brain includes the region that later, after differentiation and outgrowth from the hemisphere, receives the nerves of smell and is known as the rhinencephalon. A slight ridge (Fig. 911, 7?), projecting inward from the roof of the fore-brain, suggests a subdivision of the general space into a posterior and an anterior region. ^This use of the tenn hiud-hrain is at variance with its older sisjnificance, still retained by some German writers, as indicatinij the upper division (metencephalon) of the posterior priman' vesicle. In view, however, of the now sjeneral application of fore-brain and mid-brain to the other primarv vesicles, it seems more consistent to include hind-brain in the series, as has been done by Cunningham, with a distinct gain not only in convenience, but in avoiding terms which in their Anglicised form are at best awkward and unnecessary. io6o 111". MAX ANATOMY. The latter, the outwardly buli^inj^ palliimi or hcinisphcre-anla^c, is limited below by the optic recess, the entrance into the optic \esicle, and, farther front, by a flattened triangular elevation that marks the earliest rudiment of the corpus striatum. Th6 posterior or thalamic region extends backwartl to the mid-brain, from which it is separated by the slight external constriction and correspondilig internal ridge. During the fourth week the demarcations just noted become more definite, so that the primary anterior vesicle is imperfectly subdivided into two secondary compart- ments, the telencephalon, conxenicnth' called the eud-brain, and the dienceph- alon. Consideretl with regard to the details presented by the interior of the fore- brain, the four areas recognized by His are evident. These are (Fig. 912) the region of the pallium antl of the corpus striatum, respecti\elv abo\'e antl below in the telencephalon, and the region of the thalamus and of the hipothalamus respec- tively above and below in the diencephalon. Between the protruding hemispheres, the telencephalon is closed in front and below by a thin and narrow wall, the lamina lermiualis, which dertnes the anterior limit of the brain-tube. While the more detailed account of the further development of these regions will be given in ct)nnection with the description of the several dixisions of the brain, Fic. 912. cp Mid-braiii Mid-brain DicnceDlialoii Thalainciiccplialoti Telencephalon Pallium -Spinal cord Corpus , striatum / / Spinal cord Reconstruction of brain of human embryo of about four weeks (6.9 mm.); A, outer surface; B, inner surface; /, isthmus ; os, aperture of optic stalk ; c/i. cerebral peduncle ; cj, cervical flexure ; hf, cephalic flexure. Drawn from His model. it may be pointed out here, in a general way, that the jxillium gives rise to the con- spicuous cerebral hcmisj:)heres, which, joined below bv a common lamina, expand out- ward, upward and backward and rapidlv dwarf the other jxirts of the brain-tube which are thus gradually covered over. The striate area thickens into the corpus striatum, Avhich appears as a striking prominence on the outer and lower wall of each lateral ventricle. The latter represents a secondary extension of the original cavity of the fore-brain enclosed by the developing cerebral hemisphere, and at first is large and thin-walled and communicates by a wide opening with the remainder of the brain- vesicle. The unequal growth and thickening, which subsequently modify the surrounding walls, reduce this large aperture until it persists as the small foramen of Monro, by which the lateral ventricle communicates with the third ventricle. The latter represents what is left of the cavity of the fore-braiii and. therefore, the com- GENERAL DEVELOPMENT OF THl". BRAIN. 1061 billed contribution of the tclcnccijhalon and dienccphaion. Diirinjn the fifth week the (Hencephalon exjjands into a relatively larj^e irregular space (Fij4'. 913), whose roof and floor are thin and whose lateral walls are thickened by the masses of the dexcloping thalanii. Tlic hypothalamic rej^ion becomes the most dependent part of the fore-brain and L;i\es rise to the structures that later occupy the int<-r- peduncular space on the l)ase of the brain. The roof of the diencephalon remains thin, does not produce ner\-ous tissue antl, in conjunction with the intjrowth of the vascular pia mater, forms the \elum interpositum and its choroid j)le\uses. The pineal body and the ]:)osterior lobe of the ])ituitary body arise as outgrowths frf)m the roof and floor of the diencephalon iH'spt'ctivelv. The mid-brain, or mesencephalon, at first lartje and conspicuous on acccjunt of its elontyation and prominent position at the summit of the brain-tube, does not keep pace with the adjoining; vesicles, and in the fully formed brain is represented by the parts surrounding" the aqueduct of Sylvius. Neither does it subdivide, but, while its entire wall is converted into nervous tissue, retains its primary simplicity to a greater degree than any of the other brain-segments. The lateral and ventral walls of the mid-brain contribute the cerebral peduncles ; its roof gi\'es rise to the corpora quadrigemina ; and its cavity ])ersists as the narrow canal, the aciucduct of Svlvius, that connects the third and fourth \entricles. The posterior vesicle, the hind-brain, or rhombencephalon, the largest of the primary brain-segments, is the seat of striking changes. These include thicken- ing and sharp forward flexion of the ventro-lateral walls, in consecjuence of which the floor of the space becomes broadened out opposite the bend and assumes a lozenge- shaped outline. The hind-brain is conventionally subdi\'ided (Fig. 913) into a superior part, the metencephalon, and an inferior part, the myelencephalon. Its cavity, common to both subdivisions, persists as the fourth ventricle. The extreme upper part of the metencephalon, where it joins the mid-brain, early exhibits a constriction, which by His has been termed the isthmus rhom- bencephali and regarded as a distinct division of the brain-tube. In the fully formed brain, the isthmus corresponds to the uppermost part of the fourth ventricle, just below the Sylvian aqueduct, roofed in by the superior medullary velum that stretches between the superior cerebellar peduncles. The thickened and markedly bent ventro- lateral wall of the metencephalon gives rise to the pons Varolii, whilst in the roof of the ventricle appears a new mass of nervous tissue, the cerebellum. The myelencephalon, soon limited below by the cervical flexure, shares in the ventral thickening seen in the preceding division. Its floor and particularly its sides, the latter at the same time spreading apart, form the medulla oblongata, which below gradually tapers into the spinal cord. Its roof, in which thinness is always a prominent feature, becomes more attenuated as development proceeds and is converted into the inferior medullary velum and the tela chorioidea that close in this part of the fourth ventricle. -The subsequent invagination of this membranous portion of the ventricular roof by the pia mater brings about the production of a choroid plexus similar to that seen in the roof of the third Aentricle. From the foregoing sketch of the changes affecting the embryonic brain-tube, it is evident that the anterior and posterior primary vesicles undergo subdivision, while the mid-brain remains undivided, five secondary brain-vesicles — the telencepha- lon, the diencephalon, the mesencephalon, the metencephalon and the myelence])ha- lon — replacing the three primary ones. In consequence of the unequal growth of various parts of the cephalic segment of the neural tube, the latter becomes bent in the sagittal plane at certain points, so that, when viewed from the side, the axis of the developing human brain describes an S-like curve (Fig. 912). These flexures, to which incidental reference has been made, bring about a disturbance, for the most part temporary, in the relations of the brain-segments, which in the lower vertebrates follow in regular order along an axis practically straight. In the developing human brain, in which they are most conspicuous, there are three flexures — the cephalic, cervical, and pontine. The first of these, the cephalic flexure which appears towards the end of the second week and before the neural tube has completely closed, is primary and involves the entire head. It takes place in the region of the mid-brain and lies io62 Hl'MAN AXATOMV. Uiciicephalon Telencephalon Corpus striatum Optic recess Mesencephalon Isthmus Metencephalon -^ Myelencephaloii above the anterior end of the primary i,ajt-tiibe antl of the notochord. At first the axis ol the tore-brain Hes about at ri<^ht ans^^les with that of the rhombencephalon, ( Fig. 91 1 ) but, with the in- creasing size of the middle and anterior \esicles, the angle of the flexure becomes more acute until the long axis of the fore-brain and of the rhombencephalon are almost parallel (Fig. 912). During the fourthweek a second \entral bend, the cervical flexure, apj^ears at the lower end of the hind- brain and marks the separa- tion of the encephalic from the spinal portion of the neural tube. The cerxical flexure, which also involves the head, is most evident at the close of the fourth week, when it is almost a right angle ( Fig. 912); after this it becomes less pronounced in consequence of the elevation of the head which succeeds the period when the embryonic axis is most bent. The third flexure appears about the tifth week in the part of the metencephalon in which the pons is later developed and, hence, is termed the pontine flexure. It concerns chiefly the ventral wall, which is in consequence for a time ventrallv doubled on itself ; subsequently this flexure almost entirely disappears. In contrast to the preceding bends, this flexure is only partial and in\ol\es chiefly the ventral and only slightly the dorsal wall of the neural tube ; on the exterior of the embryo its presence is not detectable. The developmental relations of the chief parts of the fully formed brain to the embrj'onic brain-vesicles are shown in the accompanying table. Table Showing Relations of Brain- Vesicles and Their Derivatives. Ventral Dorsal zone of brain-wall Diagram showing five cerebral vesicles and dorsal and ventral zones of their wall ; based on brain of embryo of four and one-half weeks. (His.) Prlmarv Secmrnt Anterior vesicle Prosencephalon Fore-brain Secondary Segment Derivatives Cavity Telencephalon Cerebral hemispheres Olfactory lobes Corpora striata Diuncejilialon Optic thalami optic nerves and tracts Subthalamic tegmenta Ititerpeduncular structures Pineal and piiuitary bodic-s Lateral ventricles 1 ___„„ j„„. Foramina of Monro 1^°"*'*'^ Anterior part of third ventricle Posterior part of third ventricle Middlt^\csi, !•.■ Mesencephalon or Miora ■lunririccmivT .Arpicduct of Sylvius Posterior vesicle Rhombencephalon Hind-hrain Superior cerebellar pcduncL Su|)erior medullary velum Metencephalon M yelencrphalon Pons Cerebellum Fourth ventricle Medulla Inferior medullao' velum Notwithstanding the great changes in position and relation which many parts of the human brain suffer during development, chiefly in consequence of the enormous expansion of the pallium and the correspondingly large size of its commissure, the gp:neral dfa'elopmknt of the hrain. 1063 corpus callosum. tlie fuiKiaiiu-ntul rdatioiiships incliaitccl by embryology arc of such value that, e\en in thi- description of the adult origan, jri-oui)int^ of the various jxirts of the brain upon a develop- Fio. 914. Kpillialiimus Mttiithalanius Pars iiianiniillaris hypothalami mental basis is found acKan- tag'cous. Althouijh strict adherence to such a plan would be at times inconven- ient, and, therefore, will not be followed, constant refer- ence to primary relations is imperative. It will be con- venient, therefore, at this place, to call attention to the accompanying outline diagrams which illustrate the principles established by His in his epoch-making studies of the human brain. In addition to showing the five cerebral vesicles. Fie Tlialaiiiu eseticephalon Pcdunculi cerebri Isthmus -Cfrcl)fllum Poms Medulla Dorsal zone Ventral zone Rhinencephalon Pars 'optica hypothalami ]• . .1 ' 1 x-'"* • Diagram showing chief derivatives from cerebral visicles : based on 913 indicates the relative ^ brain of embryo of third month. (His.) position and extent of the two fundamental subdivisions of the lateral walls of the neural tube, the dorsal or alar and the venh-al or basal laminae, which play such important roles in the differentiation of the various parts of the brain-stem. Fig. 914 shows a later stage, in which the genetic relations of all the more important parts of the brain may be recognized. The greatest complexity is presented in the development of the derivations of the fore-brain, particularly of those which are differentiated from the diencephalon and later are found connected with the third ventricle. In order to set forth the developmental relations of the fore-brain, the following table from His, slightly modified, will be of service : Fore-Brain f Telencephalon- f Pallium C Hemisphaerium ' Corpus striatum ( Rhinencephalon Prosencephalon [DIE^ Pars optica hypothalami f Pars mammillaris hypothalami I [Thalamus ENCEPHALON Thalamencephalon Epithalamus Habenula Corpus pineale Commissura post. .Metathalamus Corpora geniculata Parts of the Brain derived from the Rhombencephalon. THE MEDULLA OBLONGATA. The medulla oblongata, sometimes called the bulb and usually designated by the convenient but indefinite name "medulla," is the direct upward prolongation of the spinal cord. It begins at the decussation of the pyramids below, about on a level with the lower border of the foramen magnum, and ends at the lower margin of the pons above and is approximately 2.5 cm. (i in) in. length. Its general form is tapering, increasing in breadth from the transverse diameter of the cord (10 mm.) below, to almost twice as much (18 mm.) above, and in the antero-posterior dimen- sion from 8-15 mm. Its long axis corresponds very closely with that of the cord and is, therefore, approximately vertical. The medulla, surrounded by the pia and arach- noid, lies behind the concave surface of the basilar portion of the occipital bone, with its dorsal surface within the vallecula between the hemispheres of the cerebellum. Superficiallv, in manv respects the medulla appears to be the direct continuation of the spinal cord. Thus, it is di\ided into lateral halves by the prolongation of the anterior and posterior median fissures ; each half is subdivided by a ventro-lateral and a dorso-lateral line of nerve-roots into tracts that seeminglv are continuations of X064 HIMAX ANATOMY. the anterior, lateral and posterior columns of the cord. This correspondence, how- ever, is incomplete and only superficial, since, as will be evident after studyiii}^ the internal structure of the medulla, the c(Hnpt)nents of the cord, both ^ray and white matter, are rearrani^ed or motlitied to such an extent that few occupy the same posi- tion in the medulla as they do in the cord. The anterior median fissure is interrupted at the lower limit oi the medulla, for a distance of from 6-7 mm. , by from five to seven robust strands of nerve-fibres that pass obliquely acros,s the furrow, interlacing- as thcV jiroceed from the tw(j siiles. These strands constitute the decussation of the pyramids ( dccussatio inramidum), whereby the s^reatcr numl)cr of the fibres of the important motor j>aths pa.ss to the oi)posite sides to 1,'^ain the lateral columns of the cord, in which they descend as the lateral pyramidal tracts. The fibres that remain uncrossed occupy the lateral por- tions of the [jyramids and, conxcry^in^ towards the median fissure, descend on either side of the latter within the anterior columns as the direct i>\ramidal tracts. The Fig. 915. Optic tract Maniniillary body Pons (basilar groove) Middle cerebellar peduncU Anterior median fis'^ , — Cerebelluni- Root-bundles of ninth ..-•^^ "W^. and tenth nerves Infundibulum Cerebral peduncle Interpeduncular space :;^- Tiigeminal nerve liddle cerebellar peduncle "^"Inferior cerebellar peduncle (Restiform body) Olivary eminence Arcuate fibres 'Pyramidal decussation Root-bundles of twelfth ner\'e y^' .Anterior roots of first spinal nerve Brain-stem viewed from in front, showing ventral aspect of medulla, pons and mid-brain. decussation varies in distinctness, sometimes the component strands being so buried within the fissure that they are scarcely evident, or e\en not at all apparent, on the surface and can be satisfactorily seen only when the lips of the groove are separated. Above the decussation the anterior median fissure increases in depth in conse- quence of the greater projection of the bounding pyramidal tracts. Its upper end, just below the inferior border of the pons, is marked by a slightly expanded triangular depression, ihe/oramefi ca'cinn. The posterior median fissure, the direct continuation of the corresponding groove on the cord, extends along onlv the lower half of the medulla, since above that limit it disapj^ears in consequence of (a) the separation and di\ergence of the dorsal tracts of the bulb, which below enclose the fissure, to form the lower lateral boundaries of the lozenge-shaped fourth ventricle (fossa rhomboidalis), and {b) the gradual backward displacement of the central canal within the closed part of the medulla until, at the lower angle of the ventricle, it opens out into that space. Each half of the medulla is superficially subdivided into three longitudinal tracts or areas by two grooves situated at some distance to the sifle of the ventral and dorsal median fissures respectively. One of these, the antero-lateral furrow, marks the line of emergence of the root-fibres of the hypoglossal nerve, which, being entirely TIIH MKDILI.A ( )HL( )N(;ATA. 1065 motor, correspond to the ventral roots of the spinal nerves with' which they are in series. The other jj^roovc, the postero-lateral furrow, continues upward in a general way the line of the tlorsal spinal root-fibres and marks the attachment (»f the fibres of the ninth, tenth anil bulbar part of the eleventh cranial nerves. Unlike the posterior root-tibres of the cord, which are exclusively sensory, those attached alon^ this sjroove of the mcnlulla are partly efferent and ])arlly arferent, the fibres belonir- injL^- to the spinal accessory beinj;- entirely motor, while those of ilu- ^dosso-pharynijeal and tin- pneunii)i;aslric inclutle both and, therefore, are mixed. The Anterior Area. — This subtlivision of the medulla, also known as the/jvw- miit includes the region Ixing between the anterior median fissure and the antero- lateral furrow. Superticialiy it appears as a slightly convex longitudinal tract, from 6-7 mm. in width, that continues upward the anterior column of the cord. Each Vn.. 916. pyramid constitutes a robust strand, which ccichrai lonex below bcginsat thedecussationand, increas- ing slightly as it ascends, above disappears within the substance of the pons. Just before its disappearance, or, strictly speak- ing, after its emergence, the pyramid is slightly contracted on account of the increased width of the bounding furrows. Its chief components being the descending motor paths formed by the cortico-spinal fibres, of which approximately four-fifths pass to the opposite side by way of the decussation to gain the lateral pyramidal tract, it is evident that only to the extent of the direct pyramidal fasciculus and, for a short distance, the anterior ground-bundle, are its constituents represented in the anterior column of the spinal cord. The fibres destined for the direct pyramidal tract, which above the decussa- tion occupy the lateral part of the pyramid, gradually converge toward the mid-line as the decussating fibres disappear, until, at the lower limit of the crossing, they lie next the median fissure, which position they retain in their further descent within the cord. The space thus afforded at the lower end of the medulla, to the outer side of the uncrossed fibres, is occupied by the prolongation of the anterior ground- bundle, which, however, soon suffers displacement as it encounters the pyramid. The ground-bundle lies at first to the outer side of the strands of decussating fibres and then behind the pyramid; higher, it is pushed backward towards the mid-line by the appearance of the inferior olive and the mesial fillet until, finally, it is continued as the posterior longitudinal fasciculus at the side of the median raphe beneath the gray matter covering the floor of the fourth ventricle. The proportion of the pyramidal fibres taking part in the motor decussation is not always the same, from 80-90 i)er cent, being the usual number. \'ary rarely all the fibres cross, with suppression of the direct pyramidal tracts — an arrangement found normally in many lower animals. On the other hand, the direct pyramidal tracts may appropriate an unusually large number of the fibres, even to 90 per cent, of the entire pyramid, the crossed tract, however, never being entirely unrepresented. Ordinarily the tracts of the two sides are approximately of equal extent, but occasion- ally they may be asymmetrical, in which case the excess of the one is offset by a corresponding diminution in the fasciculus of the opposite side (Flechsig). Pyramidal decussation Lateral pyramidal tract Direct pyramidal tract Spinal nerve Diagram showing course and decussation of cortico- spinal (pyramidal) tract ; M, medulla; P, pons; CP, cerebral peduncle; T, thalamus; C, L, caudate and lenticular nuclei ; CC, corpus callosum. lo66 HUMAN AXA'IOMV. The Lateral Area. — This rci^ion is tleliiutl on the surface by the antcro-lateral and i)(»ui()-latiiai tunows in front ami hchiiul rispectivcly, and incliKlc-s a narrow strip on the hiteral aspect of the niechilla. Helow, the tract is continuous with the lateral cohnnn of the cord, a resemblance which is, however, only suiiert'icial since within tile medulla the large crossed pyramidal tract no longer lies laterally but within the anterior area of the opposite side. The upper jjart of the lateral area is consi)icuously modified by the jjresence of an elongatecl oval prominence, the olivary eminence (oliva), produced by the imderlying cf)rrugated lamina of gray matter composing the inferior olivary nucleus. The olive measures about 13 nun. in lengtli and about half as nuich in its greatest width. Its uj)jier end, more proiuinent and slightlv broader than the lower, is separated from the inferior border of the pons by a tleej) groove, which medially joins the furrow occupied by the ]iypoglf)ssal root- fibres and laterally is continuous with a broad depressed area, ihe parao/ivary fossa, that separates the olive from the restiform body and lodges the fibres of the glosso- pharyngeal and pneumogastric ner\'cs. The demarcation of the lower tapering end of the oli\e is somewhat masketl by the antc7'ior supcrjicia/ araiate Jidres, which cover for a \ariable distance the inferior part of the olive in their course backward to gain Thalamus Median .i^eiiiculate body Inferior bracliiuni Superior colliculus Cerebral peduncle Inferior colliculus Superior cerebellar peduncle Superior medullary velum Middle cerebellar peduncle Line of attachment of roof of IV ventricle Inferior cerebellar peduncle (restiform body) Clava Tuberculum cuneatum Tuberculum Rolandi Fig. 917. Lateral geniculate body Superior brachiimi Mesial root of optic tract Anterior perforated space Optic tract - Lateral olfactory root Optic nerve Optic commissure Tuber cinereum Mammillarv body Lateral area of medulla Brain-stem viewed from the side, showiiii,' lateral aspect of medulla, pons, and mid-brain. the restiform body. The components of the lateral column of the cord traceable into the medulla — the direct cerebellar and Gowers' tract and the long paths of the lateral ground-bundle — for the most part, with the exception of the direct cerebellar tract, pass beneath or to the outer side of the olive. The superficially placed direct cere- bellar tract gradually leaves the lateral area and passes outward and backward to join the inferior cerebellar peduncle by which it reaches the cerebellum. The Posterior Area. — The posterior region of the medulla is bounded laterally by the fibres of the ninth and tenth nerves ; and mesially, in the k)wer half of the biilb, by the posterior median fissure and, in the upi:)er half, by the di\erging sides of the fourth ventricle. Below, the posterior area receives the prolongations of the tracts of GoU and of Burdach, whicli within the medulla are known as the funic- ulus gracilis and funiculus cuneatus respectively, and are separated from each other by the paramedian sulcus. Beginning with a width of about 2 mm. , the gra- cile hiniculus increases in breadth as it ascends until, just before reaching the lower end of the fourth ventricle, it expands into a well-marked swelling, the clava, about 4 mm. wide, which is caused by a subjacent accumulation of gray matter. Then, diverging from its fellow of the opposite side to bound the ventricle, after a short course it loses its identity as a distinct strand and becomes continuous with the TH1-: MKDULLA OBLONGATA. 1067 inferior cerebellar peduncle or restiform body. The expansion within the ui)per part of ihv tiiiucuhis .^raciHs, the clava, contains the nucleus gracilis ( nucleus funiculi <,Macilis), the recei)ti()n station in which the lon^ sensc^ry fibres of GoU's tract are interrupted. The triangular interval inchided between the j^racile funicuh, where these begin to diverge, correspomls to the k\rl at which the central canal of the cord ends by opening out into the lourth ventricle. A thin lamina, the obex, closes this interval and is continuous with the ventricular roof. Along the outer side of the gracile fasciculus and separated from it l)\' the ijara- median furrow, extends a second longitudinal tract, the funiculus cuneatus, which at the lower end of the medulla receives the column of Hurdach. Slightly above the lower level of the clava, the cuneate strand also exhibits an expansion, the cuneate tubercle ( tuberculiMU cincreum), that is less circumscribed, but extends farther upward than the median elevation. Beneath this prominence lies an elongated mass of grav matter, the nucleus cuneatus( nucleus funiculi cuncati), around whose cells the Ion" sensory fibres of Burdach's tract end. Still more laterally, between the roots of the ninth and tenth nerves and the cuneate strand, the posterior area of the medulla presents a third longitudinal elexa- tion, the funiculus of Rolando. The latter is caused by the increased bulk of the Fic;. 918. Inferior colliculus Cerebral peduncle Median fossa Median sulcus Middle cerebellar peduncle Acoustic striae Acoustic trivone Restiform body Attachment of ventricular roof Obex Funiculus cuneatus t'renulum Superior trochlear nerve Cerebellar peduncle Floor of fourth ventricle Fovea stiperior Eminentia teres Trigonum hypoglossi Trigonum vagi (fovea inferiox) Funiculus separans Area postrema Funiculus gracilis Lateral area Medulla and floor of fourth ventricle seen from behind, after removal of cerebellum and ventricular roof. X i54- underlying substantia gelatinosa that caps the remains of the posterior horn of gray matter, and is overlaid by a superficial sheet of white matter composed of the longi- tudinal fibres of the descending root of the trigeminal nerve. While, therefore, the tubercle of Rolando is produced by the exaggeration of gray matter represented within the spinal cord, the gracile and cuneate nuclei are new stations in which the posterior root-fibres not interrupted at lower levels end, and from which the sensory impulses collected by the cord are distributed to the cerebellum and the higher centres by neurones of the second order. The upper half of the posterior area of the medulla is modified by the presence of the fourth ventricle, the lower lateral boundary of which it largely forms, into a robust rope-like strand that diverges as it ascends. Above, it abuts against and fuses with the lateral continuation of the pons and then, bending backward, enters the overhanging cerebellum as the inferior cerebellar peduncle. This strand, also known as the restiform body f corpus restiformc), is seemingly the direct prolongation of the gracile and cuneate funiculi. Such, however, is not the case, since the fibres passing from these tracts to the cerebellum by way of the restiform body are the axones of the gracile and cuneate nuclei and, therefore, new links in the chain of conduction. io68 HUMAN ANATOMY. The inferior cerebellar peduncle is the most direct path by which the cerebellum is connected with the medulla and the spinal cord. In addition to the tracts originating in the cord and destined for the cerebellum ( the direct cerebellar and possibly part of Gowers' tract), it comprises probably fibres passing in both direc- tions; that is, from the cells within the medulla to the cerebellum, and from the cerebellar cells to the medulla. A more detailed account of these components will be given in connection with the structure of the medulla (page 1072 ). Upon close insj)ection of the surface of the medulla, the direct cerebellar tract is seen as an obliquely coursing band that at the lower level of the olive leaves the lateral area and gradually i)asses backward, over the upi)er and outer end of the Rolandic tubercle, to join the restiform body, within which it continues its jtjurney to the cerebellum. The anterior superficial arcuate fibres also enter the restiform body, after sweeping around the inferior pole of the olive, or crossing its surface, and the upper part of the funiculus of Rolando. Additional contributions, the posterior supcrjicia/ arcuate Jibres, proceed to the restiform body from the gracile and cuneate nuclei of the same side. Just before bending backward to enter the cerebellum, the restiform body is crossed by a variable number of superficial strands, the striae acusticae, that may be traced from the floor of the fourth ventricle and around the inferior peduncle to the cochlear nucleus. INTERNAL STRICTURE OF' THE MEDULLA OBLONGATA. As already j^ointed out, the correspondence between the si)inal cord and the medulla is only superficial, sections across the medulla re\ealing the presence of con- siderable masses of gray matter and important tracts of nerxe-fibres not represented Fio. 919. FlK. 92V Fig. 927 Fig. 924 Fig. 922- Fig. 921 Fig. 92cr Ventral i A) and dorsal {B) aspects of brain stem, showing levels of sections which follow. in the cord, as well as the rearrangement, modification or disappearance of spinal tracts which are prolonged into the bulb. In consequence, the medulla, even at its lower end, presents new features, and towards its upper limit varies so greatly from the cord that but slight resemblance to the latter is retained. The character- istic features displayed by transverse sections of the medulla at dillerent levels depend upon the changes induced by four chief factors : — ( i ) the decussation of the pyramids, {2) the appearance of the dorsal nuclei, (3) the production of the formatio reticularis, and (4) the opening out of the fourth ventricle. Till': MF.nn.LA ohloxc.ata. 1069 I"u;. q2. I'lniiculus Kiacilis I'uiiiculus iiiieatus Siibstaiilia ^claliiiona . of iiustcrioi" coriiu Decussating fibres Anlorior coriiu y The effects of the decussation of tlie pyramidal tracts, assiiininjj; for convenience that the latter pass from below upward, arc* conspicuous when followed in <'onsecutivc transverse sections from the spino-bulbar junction cerebrahvard. The first suggestion of the decussa- tion appears ( W^. 920 ) as strands of nerve-fibres, that pass from the field of the lateral pyramidal tract in the lateral column obliquely through the adjacent ante- rior horn of gray matter and across the bottom of the an- terior median fissure to gain the opposite anterior col- umn. At a slightly higher le\el, w here the decussation is fullv established (Fig. ' ~' ' Q2I ) the laroe strands of , '''^"'^Yerse section of medulla at level A, Fig. 919; beginniiiK of pvramidai ^:\.-: '""S Siranus 01 decussation. Weigert-Pal staining, y 5H. Preoaration made by Professor obliquely sectional fibres Spiiier. are seen cutting through the gray matter, partly filling the median fissure, and collecting on either side of the latter as the large ventral bundles which thence upward constitute the prominent pyramidal fields. In consequence of the greater space required by the pyramids, the isolated anterior horns of the gray matter, cut off by the crossing strands, and the adjacent anterior ground-bundle are displaced laterally and at first lie to the outer side of the decussated fibres. Later, the ground -bundle assumes a position behind the pyramid and eventually becomes continuous with the posterior longitudinal fasciculus (page 1 1 16). The detached anterior cornu of the gray matter is pushed outward and backward and gradually becomes broken up by and interspersed among the fibres of the formatio reticularis. The Posterior Nuclei and the Arcuate Fibres. — The robust tracts of white matter (nerve-fibres) prolonged into the graciie and cuneate funiculi from the tracts of GoU and of Burdach become inxaded by new masses of gray matter, the nucleus gracilis and cuneatus. The graciie nucleus, the first encountered, begins as a narrow area of gray Fk;. 921. matter within the correspond- ing strand, on a level with the pyramidal decussation (Fig. 921). It rapidly in- creases in bulk, until it not only invades the entire funiculus gracilis, but also joins the gray matter sur- rounding the central canal. The superficial stratum of spinal fibres gradually dimin- ishes as more and more of its components end around the cells of the graciie nucleus, until, finally, all are inter- rupted. Meanwhile the cuneate nucleus appears within the funiculus cuneatus as a dorsally directed club- shaped mass of gray mat- ter (Fig. 922) which soon becomes a prominent mottled area, sharply defined by the o\erlying stratum of Burdach fibres. The cuneate nucleus extends to a higher level than the nucleus Nucleus gracilis Funiculus cuneatus Spinal root of V nerve Substantia "3^1 gelatinosa Pyramidal cit.LUSsation Transverse section of medulla at level B, Fig. 919; pyramidal decus- sation well established ; posterior cornua are displaced laterally by posterior columns. -5'^. Preparation by Professor Spiller. 1070 111 MAN ANATONn'. gracilis ami, c\x'ii after the disapiK-arancc of the latter, contiiuies as a strikiiij^ collec- tion of gray matter beneath the tlorsal surface of the medulla, from which it is separatetl by the posterior superficial arcuate fibres. Within the upper part of the fasciculus cuneatus the gray matter becomes subdivided into two masses ( I'ig. 924), the more superficial and C(jntinuous of which is called the nucleus luucalus cxtcrnus, and the deeper and more broken one, the nuclius cuneatus inlernns. Owing to the increased bulk of the fasciculi of the posterior area occasioned by the ai)pearance and expansion of the contained nuclei, the dorsal horns of the gray matter are displaced laterally and forward, so that they come to lie on a level with the central canal. Meanwhile the posterior cornua themselves, especially the capjjing substantia gelatinosa, materially gain in bulk and now apj)ear as two club-shaped masses of grav matter that cause the tlorso-lateral projections of the Rolandic tubercles seen on the Funiculus Rracilis ■— Funiculus cuneatus // Fic. 922. Nucleus gracilis eus cuneatus Spinal root of V ner\e~ / Substantia gelalinosa4 — r ■ v£' Accessory olivary nucleus-vr Anlero-lateral ground-bundle ()nilinj4 siii)erficial pn)tul)erance and the mouth, or hi/um, looks mesially and somewhat d(jrsally. When reconstructed and \ieweil from the side ( Fijj;^. 925), the plications of the lateral and dorso-lateral surfaces display a general antero-lateral disposition. On the ventral surface the grooves radiate from the ventral border of tlie hilum (Sabin). The greatest length of the inferior olivary nucleus is from 12-15 nmi., its transverse diameter is about 6 mm., and its vertical one about f)ne millimeter less. The somewhat compressed hilum measures sagittally from (S-gmm. The ])licated lamina of gray matter comjjosing the wall of the sac is from .2-. 3 mm. in thickness and contains numerous small irregularly spherical ner\e-cells, each provided with a variable number of dendrites and an axone, embedded within a compact feltwork of neuroglia fibres. The interior of the gray sac is filled with white matter consisting of nerve-tibres that, for the most part, stream through the hihun and thus constitute the olivary pedimcle. These strands, known as the cerebello -olivary fibres, connect the cerebellar cortex with the inferior olivary nucleus and probably pass in both directions. Many fibres, the axones of the olivary neurones, issue from the hilum on the one side, cross the mid-line and, sweeping through the opposite olivary nucleus either by way of the hilum or directly traversing the gray lamina, continue their course to the restiform body and thence to the cerebellum. Other fibres originate in the cells of the cerebellar cortex and proceed in the opposite direction along the same pathway to end in relation with the cells of the inferior olivary nucleus. The further links in the chain of conduction are uncertain ; according to Kolliker it is prob- able that from some of the olivary cells, fibres pass downward into the antero-lateral ground- bundle of the cord. The accessory olivary nuclei are two irregular plate-like masses of gray matter that lie res{)ecti\'ely mesially and dorsally to the chief oli\e. The first of these, the mesial accessory olivaty nu- cleus ( nucleus olivaris accessorius inesialis) is a sagittally j)laced lamina, from lo-i i mm. in length, which lies lietween the tract of the fillet and the root-fibres of the hypoglossal nerve. It extends be- low the inferior olive and, therefore, is encountered in transverse sections at a lower level — immediately above the pyramidal decussation — than the main nucleus. According to the recon- structions of Sabin, the nucleus comi>rises three dorso- ventral columns of cells, of Fig. 926. Cerebelto-olivary strands #. - 4 vj^ "^^^ 'i>^'^-..->&- ,^ Section of inferior olivary nucleus, showinj; plicated sheet of gray substance traversed by strands of cerebello-olivary fibres. X loo. THE MEDULLA OBLONGATA. 1073 which the lower and middle are continuous and the upper is unconnected, and four small isolated masses of j^ray matter alonj^ the dorsal border of the nucleus. The inferior or spinal end of the nucleus is thickened and bent outward, so that its plane is oblique and parallel with the ventral surface of the chief olive. Higher, when the latter is well established, the mesial accessory nucleus is represented by a narrow broken tract, that corresjx)nds more closely with the sagittal plane. In this situa- tion the nucleus lies between the fillet and the inner end of the chief olive antl across Dorsal nucleus of vagus Fig. 927. Ventricular roof Funiculus cunea- tus, overlaid by restiform body Substantia geKi tinosaoverla by root of \ Nucleus ambiguuB Nucleus cuneatus .poglossal 4L, Post, longitudinal *' ^J/ "fasciculus Nucleus lateralis- Root-fibres of XII Interior olivary nucleus Tract of mesial fillet Pyramidal tract Anterior superficial arcuate fibres Transverse section of medulla at level E, Fig. 919 ; central canal has opened into fourth ventricle ; restiform body appearing. X 5. Preparation by Professor Spiller. its hilum. The dorsal accessory olivary nucleus (nucleus olivaris accessorius dorsalis) is less extensive than the median, measuring about 9 mm. in length, and lies close to and behind the posterior lip of the hilum of the inferior olive. The Central Gray Matter. — As pointed out, within the closed part of the medulla the central canal and the surrounding gray matter are gradually displaced dorsally in consequence of the increasing space required by the pyramid, the fillet tract and the posterior longitudinal fasciculus, three paired tracts of longitudinally coursing fibres that lie close to the median raphe and enlarge as they are followed upward. When the central canal opens out into the fourth ventricle, the sur- rounding gray matter is correspondingly spread out and forms the lining of the ventricular floor. Within this gray sheet and near the mid-line, on each side, is seen the group of cells constituting the hypoglossal nucleus from which the fibres of the twelfth cranial nerve arise. These strands take a direct ventro-lateral course through the medulla and emerge on the surface in the groove between the pyramid and olivary eminence. Slightly more lateral, and to the outer side of the hypoglossal nucleus, another group of cells marks the position of the elongated vago-glosso- pharyngeal nucleus, partly sensory and partly motor, belonging to the tenth and ninth cranial nerves. The fibres of the vagus traverse the medulla laterally and meet the surface at the junction of the lateral and posterior areas. In this way the di^'erging fibres of the tenth and twelfth nerves subdivide each half of the medulla into three triangular areas — a mesial, a lateral and a posterior (Flechsig). Viewed in transverse sections through the upper third of the medulla, the poste- rior area — the space between the \'agus fibres and the dorsal surface of the medulla — is seen to contain a number of important fibre-tracts. ( i ) The restiform body appears 68 1074 Hl.MAN ANATOMY. as a large irregularly cresceiitic tract of transversely cut fibres that occupies the greater part of the periphery. (2) The descending root of the vestibular net~'e is seen to the inner side of the dorso-mesial border of the restiform body as a field of loosely grouix-d bundles of cross-sectioned ner\e-fibres. (3; The fasciculus solitarius, or Fig. 92S. Ventricular roof Fasciculus solitarius Dorsal nucleus oi X I Nucleus of XII I Post. I'-.nif. fasi ; u Restiform bod\ • Restiform body Root-fibres of X nerve ' Nucleus ambiguus^ — Root-fibres of .\ 1 1 ' Descending ~ vestibular root Gray column of vestibular root -.- " — Form. ret icRTisea ^ Form, retic. alba Interolivarj' stratum I median fillet) Inferior olivar\- nucleus ^\J r; Pyramidal tracts Transverse section of medulla at level F, Fig. 919; ventricular floor is wide; restiform body well established ; descending root of vestibular ner\e is seen. ■ 5. Preparation by Professor Spiller. descending root of the vagus and glosso-pharyngeal ner\es, shows as a conspicuous transversely cut bundle which lies ventro-mesially to the vestibular root. (4) The descending root of the trigeminal nerve is easily identified as a superficial crescentic field that on its mesial aspect encloses the remains of the substantia gelatinosa Rolandi. The lateral area, between the diverging vagus and hypoglossal root-fibres, is chiefly occupied, in addition to (i) the iyiferior olivary and (2) dorsal accessory olivary nucleus, by the feltvvork of fibres producing the reticular formation. In con- trast to that within the ^^' 9^9- anterior area, the retic- ulum within the lateral area contains a con- siderable amount of diftuse gray matter be- tween its fibres, and, hence, is known as (3) X.\\^ format io reticularis grisea. Accessions to the irregularly distrib- uted nerve-cells occur as two moredefinitecol- lections ; one of these, C4) the nucleus am- biguus. consists of an inconspicuous group of large cells lying about the middle of the grav Nerve-cell >:> •'f> z' '^ Longitudinal fibres* r^T'- •€ Portion of fonnatio reticularis grisea. showing nerv'e-cells and interlacing transverse and longitudinal fibres. X 130. reticular substance and is of importance as the nucleus of origin of at least part of the motor fibres of the vagus ner\'e. The other (5 ). the nticlcus lateralis, includes an uncertain aggregation of medium sized cells, situated nea:r the peripher>' and ventral TMK MKDl'LLA OHLONCiATA. lO /3 from the trigeminal root. A separate group of somewhat hirger cells, nearer the ventral border of the trifacial root, has been designated the nucleus lakralis dorsalis, and by K()lliker regarded as belonging to the origin of the spinal accessory nerve. Cochlear fibres crossing restifuriii body Descendint; root of vestibular nerve Fig. 930. V"^' Kestiform body Median Strix Nucleus vestibular irii«tit.Te of I.K nucleus IJeiters' nucleus I'ost. lonif. fasciculus Root of IX nerve Substantia .£;elati nosa Forinatio reticularis alba '"j Tract of mesial fillet — V" Inferior olivary body ^^^ :.■ \ \ Fibres of IX nerve V \ Spinal root of V neive ^ Substantia gelatinosa I'omiatio reticularis grisea \ Pyramidal tracts Tia:isverse section of medulla at level G, Fig. 919; \eiitr.U part is narro.ver, whilst dorsal part is expanded owing to increased size of restitoim Dodies. X 4. 'Preparation by Professor Spiller. In a general way the cells of these nuclei (ambiguus and lateralis) of the substantia grisea may be regarded as the analogues of the lateral horn-cells of the cord, just as those of the hypoglossal nucleus resemble the anterior root-cells of the spinal nerves. The anterior area, between the mid-line and the hypoglossal root-fibres, is occupied ventrally by ( i ) the pyramidal tract, which appropriates the entire width of the field with the exception of a very narrow peripheral zone that intervenes Fig. 931. %^ Nerve- Longitudinal Transverse { I cell fibres fibres ' ,. Median raphe Portion of transverse section of medulla, showing median raphe and adjacent formatio reticularis alba. X 130- between the pyramidal fibres and the surface along the median fissure and the ventral aspect of the medulla. This zone is traversed by (2) the anterior superficial arcuate fibres, among which is lodged an irregular column of nerve-cells that constitute (3) I076 HUMAN ANATOMY. the arcuate nucleus. The latter lies at first chiefly on the ventral and. hij^her, on the mesial aspect tjf the pyramidal tract. The cells of this nucleus, small and fusiform, are the oritjin of not a few of the superficial arcuate fibres, although those from the dorsal nuclei continue their course over the nucleus without interrui)tion. At the upper end of the medulla, the cells of the arcuate nucleus increase in number and mingle with those of the nucleus of the raphe and the pontine nucleus. Dorsal to the pyramid and immetliatcly next the mid-line lies (4) the compact tract of the median Ji/hi, composed of longitudinal fibres that are the upward continu- ation of the deep arcuate fibres, which, from the sensory decussation to the upper limit of the cuneate nucleus, bend sharply brainward after crossing the mid-line. Ihe fillet-tracts are also known as the interolivary stratum, as they constitute a compact and laterally compressed field between the inferior olivar}- nuclei. Lateral to the fillet, between the latter and the hypoglossal fibres, lies (5) the mesial accessory olivary nucleus. (6) 'Y\\q posterior longitudinal fasciculus appears in cross-section as a compact oval or laterally flattened strand, which lies next the raphe and immediately beneath the gray matter covering the floor of the fourth ventricle. This important path will be later described (page 11 16). The remaining space of the anterior compartment, between the pyramid and the ventricular gray matter, is occupied by the formatio reticularis alba, so designated in distinction to the formatio grisea on account of its meagre number of nerve-cells, since, with the excep- tion of those scattered in the immediate vicinity of the mid-line > nucleus raphe), few cells are present. The Formatio Reticularis. — Repeated mention has been made of the reticu- lar formation jjroduced by the interwea\ing of the horizontal and vertical fibres. Whilst particularly conspicuous within the medulla at the levels occupied by the gracile, cuneate and inferior olivary nuclei, on account of the prominence of the arcuate and cerebello-olivary fibres, the formatio reticularis does not end with the disappearance of these nuclei and fibres, but is prolonged upward, although less marked, by transversely coursing fibres derived from the reception-nuclei of various cranial nerves — the vagus, glosso-pharyngeal. auditory, facial, and trigeminal — from whose neurones axones of the second order arise that sweep across the mid-line to join chiefly the fillet tract or to end, perhaps, about nerve-cells of other nuclei. In this manner the formatio reticularis finds representation within the dorsal or tegmental areas of the pons and the cerebral crura. The longitudinal fibres within the formatio reticularis grisea are derived from many sources. Some are the continuation of Cowers' tract ; some belong to the long strands concerned in establishing reflex paths connecting the corpora quadrigemina. nucleus rubrum. vestibular and olivary nuclei with the spinal cord ; some are the axones of tegmental neurones and pursue shorter courses, both descending and ascending, as association fibres linking together different le\els of the brain-stem : while still others are the prolongations of the spino- thalamic and other long tracts of the antero-lateral ground- bundle of the cord. The longitudinal fibres of the formatio alba are chiefly the components of the mesial fillet and of the posterior longitudinal fasciculus with, possibly, the addition of short association fibres proceeding from the nerve-cells that are found within the anterior area. The details of a transverse section passing just beneath the lower border of the pons (Fig. 932) var>- considerably from those of the level shown in Fig. 930. The ventral half of the medulla has lost in width in consequence of the disappearance of the superficial olivary emi- nence, the inferior olive being at this level represented by only a few irregular plications. The pyramids, likewise, are narrower, and separated by the broadened anterior median fissure. The mesial fillet and the posterior longitudinal fasciculus are now widely separated by the inter- vening nucleus centralis inferior that appears between them along the raphe. The nuclei of the hypoglossal and glosso-phar>ngeal nerves are no longer seen, but instead, along the floor of the ventricle underlying the area acustica. appears a large triangular ma.ss of gray matter, the mesial vestibular nucleus. Fxtemal to the latter the lateral or Deiters' nucleus and the descending or spinal acoustic root lie close to the restiform body, which in transverse section presents a bean-shaped outline. Between the restiform body and the descending trigeminal root, the fibres of the mesial or vestibular part of the auditor)- nerve pass backward to gain the vestib- ular nuclei. The outer surface of the restiform body is closely related to a considerable THE PONS VAROLII. 1077 tract of sray matter that collectively constitutes the reception-nucleus of the cochlear division of the auditory nerve. This j^anglion is subdivided into a superior and an inferior portion, these bein^ the dorsa/ corh/tar iiiu/ius and the ventral lOihltar nin /tus resiK-ctively. They both receive the fibres of the cochlear or lateral division of the auditory nerve. The ventral cochkar nucleus is the startin.n point of a tract of transverse fibres, that pass horizontally inward, many traversing the fillet and crossing the raphe, and intermingle with those from the ()ii|)osite side. They thus form a broad strand, the lorpiis (rapczoidts, that within the pons occupies the lower limit of the tegmental region, which it separates from the ventral. In Fig. 932 Fig. 932. Mesial vesiil>iil,ir ntsciml- Gr.iy sulistaiice nucleus Deiters' nucleus Sulistanti.i ing root of floor .Nucleus of Post, lonif . / A ./ — - >;elatinosa ofVIlI ofventricle facial fasciculus '— <^ /'". N'estilnilar nerve Restiforni ISi^-' / -/ / "V:^- ■ Kestiroriii l».ly »^ ^Cotlilr:ir ner%e 1/ ■■■^m ,1 ■ ■.f ■'.■ V'<-i )orso-liter:il Superior Ventral cochlear nucleus olive Cochlear nerve and ventral cochlear nucleus / Spinal root of V / y ^>f'' Trapezoidal fibres / /Y- Inferior olivary nucleus / \.^' ^-' Median fillet \l , ' Pyramiilal tract ' Transverse section of medulla at level H, Fig. 919 ; pyramids are small and inferior olivary nuclei are disappearing; roots of auditorj' nerve are entering in relation to resliform bodies. X 4- Preparation by Professor Spiller. only the beginning of this tract is visible, but slightly higher, in the pons (Fig. 933), the trapezoidal fibres are shown in force. .Strands of fibres from the cochlear nuclei arch over the restiform body and proceed beneath the ventricular floor to the mid-groove ; these mark the course of the stritr acusticcr seen crossing the ventricle. Ventro-mesial to the spinal root of the trigeminus and the associated Rolandic substance the nucleus of the jacial nerve appears as an irregularly oval and somewhat broken group of large stellate cells, from which the strands of root-fibres pass dorso-medially. THE PONS VAROLII. Viewed from in front, the pons appears as a quadrilateral prominence on the ventral aspect of the brain, interposed between the medulla oblong^ata below, the cerebral peduncles above, and the cerebellar hemispheres at the sides. Its lower and upper limits are well defined by grooves that separate the corresponding borders from the adjacent divisions of the brain -stem, and between these boundaries the pons measures from 25-28 mm. in the mid-line. Laterally, however, its limits are unmarked, as here the mass of the pons narrows and is directly continued on each side as a robust arm which sweeps downward and backward into the cerebelhnn as the middle cerebellar peduncle. The fibres of the trigeminal nerves, which are attached near its upper and lateral margins, are taken as the conventional lateral limits of the pons, the transverse diameter measured between these points being about 30 mm. The ventral surface of the pons, strongly convex transversely and less so in the opposite direction, lies behind the basilar process of the occipital bone and the dorsum sellae. It is marked by a shallow median groove (sulcus basilaris), which broadens as it ascends and lodges the basilar artery and is bounded on each side by a slight longitudinal elevadon. Where the latter meets the medulla, the pyramid is seen to plunge into the pons beneath its transversely striated surface. The longitudinal 1078 HUMAN ANATOMY. ridges are produced by the uiulerlyinjn pyramidal tracts in tlieir journey through the pons from the cerebral peiluncles to the medulla. The trans\erse striation indicates the general course of the superficial fibres towards the cerebelhmi. The lateral surface, continued from the ventral without interruption, above is rounded and sloping and .separated from the cerebral jieduncles by a distinct furrow. Below, it passes insensibly into the middle cerebellar peduncle, into which the lower and lateral part of the pons is prolonged. Whilst the superficial striation in a general way follows the contour of the pons, a broad band ( fasciculus ol(lii|uus pontis) from the upper part of the \entral surface sweeps-oblicjuely backward and downward antl overlies the mon- horizonlally ilirected middle and lower fibres. The free portion of the dorsal surface of the ])ons contributes the upper half of the floor of the fourth \entricle and is, therefore, not visible until the roof of that cavity is removed. Above the middle j^eduncle, the sides of the pons are blended with the overlying superior cerebellar peduncles, which, in conjunction with the intervening; superior viedtillary velum, complete dorsally the ring of tissue sur- rounding the narrowed superior end of the fourth ventricle. INTERN.M. STRICTURE OF THE PONS VAROLII. \'iewed in transverse sections the pons is seen to include two clearly defined areas, the ventral and the dorsal (Fig. 933). The ventral part (pars l)asilaris ) presents a characteristic picture in which the large j^yraniidal tracts are covered in Y\o. 933. Abducent Rbres Superior cerebellar lieduntle ~ Facial fibres Substantia (.'clatinosa Spinal root of \^ Facial nucleus Trapezoidal fibres SuDerior cerebellar peduncle Nucleus Post, long Nucleus of VI fasciculus of VI -; merging facial fibres . \'estibular fibres -Spinal root of \' -Olivary peduncle ; i — ^ Superior olive Superior olive' •4." < ^^iv: rormatio reticularis ofteKmentJm Pyramidal tracts ~^. 1 ransversc fibres V Transverse section of pons at level I. Fig. 919; showing; g:eneral subdivision into ventral and dorsal tegmental) areas and nuclei of sixth and seventh nerves. \ 3. and excluded from the surface bv a conspicuous layer of superficial transverse fibres (stratum supcrticialo pontis ), that laterally sweep backward into the cerebellar peduncle and are traversed by the root-fibres of the seventh and eighth nerves. The pyra- mids no longer appear as compact fields, but are broken up into smaller bundles by the transverse strands of ponto-cerebellar fibres. This subdivision becomes more marked at higher levels of the pons (Fig. 936), in which the interweaving of the longitudinal and transverse bundles produces a coarse feltwork (stratum comploxum ). At the upper border of the j)ons, the scattered pyramidal bundles become once more collected into two compact strands, which are continued into the central part of the crusta of the cerebral peduncle. The dorsal hmit of the ventral field is occupied by a well marked deeper layer of transverse fibres (stratum profundum pontis). A considerable amount of gray matter, collectively known as the pontine nucleus Tin: PONS VAROLII. ,079 (nucleus pontis ) is distributed within the interstices between the bundles of nerve- tibres. The cells of this luiclc-us, small in size and stellate in form, are closely related to the pontocerebellar fibres of the same and of the opposite side, many constituting' stations of interruption in the cortico-cerebellar paths. The dorsal or tegmental part of the pons (pars dorsalis pontis) resembles to a considerable extent in its L^eneral structure the formatio reticularis j^risea of the medulla, consisting for the most part of a reticulum oi transverse and longitudinal fibres, interspersed with ner\e-cells, on each side of the median raphe. 'l"he appear ance of certain new masses of ^ray matter and of nerve- i*'tJ- 934- fibres, toi^c'ther with chanijes in the position of the fillet, :. _; produce details that \ary ' ..__ w-ith the level of the section. r^^'" ? ;. When this passes above the r. . -« 'y lower mary;^in of the pons -f .' (Fig- 933). two diveri^ing ^ ' :^' • ^--^r*- and obliquely cut strands of fibres, coursing from the ^•entricular floor towards the ventral aspect, mark the root-' fibres of the sixth and seventh cranial nerves and divide the dorsal region, on each side, into three areas. The middle - '^";^~ - -fi: -^ ■'/ area, between the abducent ---S^^^^fe^^ fibres mesially and the facial --^ssSi^j^--- ---- fibres laterally, contains three important collections of nerve- ^°'*'°" °^ '■''°'"'^^*iZrius°"': 'S"'"*^ ^^"' °^ ^""^'""^ cells. One of these, the nu- cleus of the sixth nerve, lies close to the floor of the ventricle and beneath the rounded prominence of the eminentia teres, which it helps to produce, and gives origin to the root-fibres of the abducent nerve. These fibres take an obliquely ventral path, slightly bowed towards the raphe, and cut through not only the dors;il but also the ventral part of the pons to gain its low-er border, along which they emerge a few millimeters from the mid-line. In favorable sections the nucleus of the sixth is seen separated from the floor of the fourth \entricle by the arching fibres of the facial ner\-e. Another conspicuous nucleus of the middle area, the superior olive ( nucleus olivaris superior ), lies near the ventral limit of the tegmental area, partly lodged within an indentation on the dorsal surface of the conspicuous tract of transverse fibres, known as the corpus trapezoides, that extends from the ventral cochlear nucleus medially and materially aids in defining the ventral boundary of the dorsal area. The superior olive (Fig. 933) is an irregularly spherical collection of nerve-cells, interposed in the path connecting the auditory nuclei with the cerebral cortex, and closely related with the tract of the lateral fillet (page 1082 ). In addition to contrib- uting numerous fibres to the latter, the superior olive sends others to the abducent nucleus which are seen as delicate strands, the peduncle of the superior olive, that pass towards the nucleus of the sixth nerve and bring this centre into relation with auditory impulses. A small collection of nerve-cells between the fibres of the trape- zoidal tract, ventro-medial to the superior olive, constitutes the nuc/eiis trapezoides. Close to the medial border of the superior olive a small oval bundle of longitudinal fibres, the central tegmental fascicnbis, is sometimes seen. These fibres are probably derived from the olivary nucleus (Obersteiner). The facial nucleus, a conspicuous but broken o\al mass of gray matter (Fig. 933). includes sexeral groups of large stellate cells that lie dorso-lateral to the superior olive and to the innei* side of the emerging facial fibres. From the cells of this nucleus the loosely collected root-fibres of the facial nerve pass back- ward and inward to reach the floor of the fourth ventricle. Here they converge into io8o lllMAX ANATOMY. a compact strand that, as the ascciidinj^ portion of the nerve, courses beneath the eniinentia teres seen on the ventricular Hoor, close to the mid-line, until it bends outward and, archinj^ around tin- abducent nucleus, continues \cntrally as the emerj^nn^^ root-fibres. The \entral j)art of the inner area and the adjoininj,^ part of the middle one are occu])ied by the field ol the mesial fillet which, at the le\el under vonsideration, no lonjrer has its longest a.xis directed dorso-ventrally, but api)ro,ximately horizontal. The tract now appears as a modified oval, somewhat compressed from before f)ack- ward, the thicker inner end of which reaches the raj)he while the tapering outer end lies near the su))erior oli\e. The posterior lotii^itudinal fasciculus is seen as a com- pact strand, immediately beneath the gray matter of the ventricular floor and at the side of the rajjhe. To the outer side of the emerging facial fibres, and therefore in I'lG. 935- /:.v^' Superior cerebellar peduncle / ' ;j;\' '.Iffif ''^ ■ * Inferior cerebellar peduncle '-'At m"Mt>. ' ,.>v. Mesencephalic root of V Posterior longitudinal fasciculus Sensory irigemina! nucleus -~ Middle cerebellar peduncle jiv^Motor trigeminal Tnucleus .! -^ Motor fibres of V -T^Trigeniinal nerve Superior olive Median fillet Deep transverse pontine fibres midal tracts ransvcrse pontine fibres Transverse section of pons at level J, Fig. 919, showing root of trigeminal nerve with its nuclei. X 3. Preparation by Professor Spiller. the lateral pontine area, appear the substautia gclatinosa and the associated spinal root of the trigeminal nerve. Just behind the latter the dcscetidiiig vestibular root lies close to the inner side of the restiform body. The collection of nerve-cells marking Deiters' nucleus is seen beneath the ventricular floor in close relation with the descending vestibular root. Sections passing at the level f)f Fij^. 935, and, therefore, about three millimeters above that of Fig. 933, show interesting details connected w ith the nuclei and roots of the trigeminal nerve. At this level the nuclei and roots of the sixth and seventii nerves are no longer seen. The median fillet appears on each side as a compressed oval, the long axis of which is hori- zontal and whose inner end almost touches the raphe. Just above the outer end of the fillet, the cerebral extremity of the superior olive is still visible, to which a few strands of transverse fibres — the last of the trapezoid body — pass. The lateral boundary of the ventral part of the pons is defined by a hugh tract of obIi()uely cut fibres that marks the entering sensory root of the trigeminal nerve. On following this tract dorsally it is seen to enter a large mass of gray matter, the sensory nucleus of the trigeminal nerve. This ganglion, composed of closely packed small multipolar cells, corresponds to an accumulation of the substantia gelatinosa, which, it will lie remembered, is to be seen in all the preceding lower levels intimately related thp: pons varolii. 1081 to the di'sccndinc; or spinal root of tlie fifth nerve. A second and more compact gangUon, the motor nucleus of the trigeminus, Ues to the inner side and siiglitly fartlier back. It contains large nuikipolar cells, extends to a somewhat hi},'iier level than the sensory nucleus, and is separated from the latter by a strand of tibres uiiich arch over the motor nucleus and then pass mesially beneath the ventricular floor to the raphe, where they cross to the motor nucleus of the