A CONTRIBUTION TO THE PATHOLOGY OF TRAUMATIC EPILEPSY Comprising the Report of the Microscopical Examination in Two Cases Operated upon by Trephining BY IRA VAN GIESON, M.D. ASSISTANT IN HISTOLOGY AT THE COLLEGE OF PHYSICIANS AND SURGEONS COLUMBIA COLLEGE, AND PATHOLOGIST TO THE CITY HOSPITAL, NEW YORK Reprinted from the Medical Record, April 29, 1893 NEW YORK TROW DIRECTORY, PRINTING AND BOOKBINDING CO 201-213 East Twelfth Street 1893 A CONTRIBUTION Pathology of Traumatic Epilepsy, TO THE Comprising the Report of the Microscopical Exam- ination in Two Cases Operated upon by Trephin- ing. For the opportunity of presenting these cases I am in- debted to Professor Starr, who gave me the clinical his- tories, and the portions of the brains removed at the operations for microscopical study. The impetus which localization has given to brain surgery, and its rather ex- tensive application at the present time, lends some interest to the study of these specimens. But attention is more especially directed to the question as to whether certain minute, very delicate changes, rather difficult to recognize in the removed fragments of the motor zones in these cases, may be considered as underlying the phenomena of epilepsy. The clinical histories are as follows: Case I. Trauma—General Convulsions Beginning in Left Arm—Splinter of Bone in the Brain Removed—Re- covery—Recurrence of Fits—Death.—A. G , male, aged twenty-four, met with an injury in April, 1888, which produced a fracture of the skull on the right side, at about the middle of the coronal suture. After the in- jury he was ill with fever and delirium about six weeks, but gradually recovered. Three years after this injury he began to have convulsions, from which he had suffered at intervals up to April, 1892, when he was first seen. The attacks began with a movement of the left arm, and a sen- sation of numbness in the left hand, and with a turning of the head to the left; he then lost consciousness and the convulsion became general. He has had as many as 4 A CONTRIBUTION TO THE two fits in a day, and the longest interval during the year was nine weeks. He had three fits in March, 1892. He was very dull mentally, and had been treated with very large doses of bromide of potassium, which diminished the frequency of, but did not arrest, the fits. Operation by trephining was performed by Dr. Mc- Burney on April 2, 1892. The skull was opened at the point of fracture over the arm centre on the right side. The external table was found to be fractured, but the in- ternal table appeared to be uninjured. The dura was very much thickened, and the pia and brain were decid- edly oedematous and yellower than normal. The pulsa- tion in the brain was greater around the softened discol- ored area than in it. This discolored area pitted upon pressure, and to the touch gave the impression as if a cyst lay beneath, but puncture in all directions with a hypo- dermic needle failed to reach any cyst. The wound healed easily. He had no paralysis, and in three weeks he was discharged from the hospital. At that time he had very much improved mentally, and had had no fits. Soon after leaving the hospital the fits began again, and in the summer they occurred with greater frequency than before the operation, and in August he died in convul- sions. Case 11. Trauma—Spasms of Right Hand—Cyst Re- moved—Recovery for Six Months—Recurrence—Second Trephining—Recovery.—Male, aged fourteen, at the age of four had a severe fall, fracturing his skull over the left coronal suture. As a result of this he developed right hemiplegia with partial right hemianaesthesia, but with- out any aphasia. Traces of this hemiplegia still remain. At the age of twelve and a half he had a second fall, hit upon his head, and soon after this he began to suffer from Jacksonian epilepsy. His fits always began with a tingling and spasm in the right hand, which extended to the arm and then down the right leg, the face being very rarely involved, though occasionally the head turned to the right. There was no loss of consciousness during the attack. It lasted about a minute, and he felt slightly weaker in the arm and leg after it. He has had as many as six attacks in a day. The boy was mentally very bright and PATHOLOGY OF TRAUMATIC EPILEPSY. 5 had no headache. Evidence of an old depressed fract- ure was found in the skull, the depression extending for- ward over the first frontal convolution, so that its posi- tion was decidedly anterior to the motor area of the arm. Medical treatment having failed to relieve these attacks, it was resolved to trephine. The point selected was the arm centre in the upper third of the central convolu- tion, though its position was an inch and a half posterior to the position of the old fracture. Dr. McEurney op- erated at Roosevelt Hospital on January 30, 1892. On exposing the dura it was found adherent to the bone and did not pulsate. When the dura was laid back it was found adherent to the pia, which was thickened and opaque so that the brain was not visible beneath it. On dividing the pia a cyst was found lying beneath the sur- face of the brain, and from this a drachm of clear fluid was evacuated. The cyst had lain in the pia itself. The walls of the cyst were removed. A strand of thickened pia was found running forward toward the old scar. The opening in the bone was therefore enlarged in the direc- tion of the old fracture until this was reached, and a sec- ond cyst was found beneath the old fracture. This cyst was also evacuated of about two drachms of fluid and its walls taken away. The brain beneath the cysts appeared to be somewhat atrophied but pulsated normally. It had an appearance of being slightly more yellow than normal brain-tissue, and the number of blood-vessels and capil- laries over its surface seemed to be rather increased. The wound was closed and healed well, and from January 30, 1892, the date of operation, until April, the boy had no fits at all. He then returned to the clinic, complaining of a return of his old attacks. On examination of the head it was found that there was a small collection of pus beneath the scalp, over the site of the opening in the bone. This pus was evacuated and the small abscess-cavity at once healed. From that date until August, 1892, the boy had no attacks. Then his attacks began again, and increased in frequency until in December he was having three or four daily. These attacks began with tingling and twitch- ing in the right hand which extended up the arm and shoulder, then down the side to the leg, arm and leg twitch- 6 A CONTRIBUTION TO THE ing together for the space of from five to fifteen minutes. Subsequently to the attacks both arm and leg were slightly paretic, the face never being involved, and con- sciousness not being lost. The use of bromides during this period had no effect upon the increase of the attacks, and he was therefore again advised to go into the hospital for operation. On January 7, 1893, Dr. Mcßurney op- erated. On exposure of the shaven head the scalp was seen to be thick and tense, so that at no place was there any perceptible depression around the old scar or over the defect in the bone. Pulsation of the brain was perceptible by palpation over the area from which the bone had pre- viously been removed, and which corresponded to the arm centre. The tissues were very much thickened, and it was thought best to avoid their direct incision. A semilunar incision was therefore made, the summit of which passed somewhat more to the left of the median line than the preceding incision and by dissecting up its anterior and posterior portions, the healthy bone below the old trephine was reached, the scalp being carefully dissected away from the old scar-tissue. A triangular opening was then chiselled in the bone about an inch and a half long and three-fourths of an inch wide. The bone was found to be closely adherent to the dura. The dura was seen to be thickened, and on being divided and turned back it was closely adherent to the pia. The pia and brain were found to be welded together in a thick connective-tissue mass. Palpation of this gave the im pression of fluid beneath it. Puncture with a hypoder- mic syringe brought away a small amount of clear serous fluid from a cavity about half an inch beneath the cortex. Incision was made into this cavity through the brain above it. When the brain-tissue was incised it was found to present an abnormal appearance. There was no clear line of demarcation between the cortex and the white matter beneath it, but a connective-tissue mass had taken the place of the cortex. This mass of tissue was there- fore excised, a piece of a lens shape, about an inch long by half an inch wide, being removed. It appeared to be scar-tissue. The second puncture with a hypodermic needle, at a point an inch farther forward, revealed the PATHOLOGY OF TRAUMATIC EPILEPSY. 7 presence of another cyst, and the incision in the brain was therefore carried forward so as to empty this. Hem- orrhage was pretty free, but after the scar-tissue had been excised the sides of the wound in the brain was seen to consist of fairly normal gray and white substance. The wound was packed with iodoform gauze and dressed anti- septically. The next day the boy was very comfortable, had no paralysis or anaesthesia. Within two weeks the wound had healed. He has had no attacks up to March, 1893. Microscopical Examination. —ln describing these morphological changes in the motor cortex, which harmo- nize very well with the symptoms of epilepsy, it is of es- pecial importance to preface the details of the examina- tion with some general remarks about the technical limitations of investigations in the finer pathology of the cortex, and the extreme difficulties of detecting and attach- ing significance to the very early and subtle changes in the cortical elements. In such a preface the investigator should indicate the great caution and most refined tech- nique which a study of minute cortical changes demands ; for then the reader will appreciate that the observer has guarded against mistaking for lesions entirely artificial changes, or normal structures which, especially in the cortex, are by no means easy to define. The difficulty in the way of research in cortical pathol- ogy is the complexity of the brain cortex; it is most su- premely highly organized, and is far beyond all other organs and tissues in the textural delicacy of its anatomi- cal elements and complexity of their arrangement. In most of the other organs the structure of the parenchyma is comparatively simple, and the stroma is arranged in such a way that there is a contrast between the two in the sections; thus in the kidney or liver, for example, the changes in the stroma or in the parenchyma attend- ing a chronic inflammation may be determined very ac- curately. The stroma is so distinct from the parenchyma, and its distribution is so readily followed, that a very be- ginning of an increase in its substance may usually be easily and positively recognized. In the same way the distinctive distribution of the comparatively simple par- 8 A CONTRIBUTION TO THE enchyma cells permits early changes in them to be deter- mined with but little difficulty. When we come to the brain cortex, however, the con- trast between stroma and parenchyma, which in other organs affords most valuable topographical aid, is lost, and the determination of changes in either stroma or parenchyma is correspondingly difficult. For in the brain cortex the neuroglia and ganglion cells, correspond- ing respectively to the stroma and parenchyma of other organs, are not only more intricately constructed, but are diffusely arranged. The neuroglia and ganglion cells are mingled together in a most intricate way, and are sur- rounded by a great wilderness of processes derived from both, which forms a very large part of what is conven- iently called the basement substance of the gray matter. Thus it can be understood what a difficult matter it is to determine any beginning increase or proliferation of the neuroglia, which in ordinarily stained sections pre- sents itself as multitudes of small round nuclei scattered all through the gray matter, without any boundaries or limitations. This problem of the determination of a very early increase in the neuroglia becomes the more baffling because, as a rule, this tissue grows so slowly that the all- important criterion of the proliferation of cells, namely, the phases of karyokinesis, are difficult to find. The investigation of minute and early changes in the other intrinsic element of the cortex—the ganglion cell —is rendered difficult by the presence of artifacts or ar- tificial changes occurring after death. The structure of the ganglion cell is so delicate and intricate, and the cor- tex is so slowly permeable to the bichromate solutions, that a number of post-mortem changes are liable to occur in the cell or are induced by the action of the hardening agents. Such artificial changes may simulate very close- ly the results of disease, and when these artificial changes are present in a cortex with suspected disease of the gan- glion cells, it becomes exceedingly difficult to understand the lesions, or to determine in what degree the changes are due to disease and in what degree to artificial condi- tions. With the best of care we can recognize, after all, but the PATHOLOGY OF TRAUMATIC EPILEPSY. 9 coarser and grosser lesion in the ganglion-cell body, which is only a part of the cell. Changes in the great forest of processes of the cell, representing a volume of protoplasm fully as large, if not larger, than the cell body itself, are beyond our cognizance even with Golgi’s methods, which seem to be of little service in showing minute changes in the ganglion cells. The aid of mitosis as an index of pathological changes in the ganglion cells is also absent, since the latest studies on this subject show that the ganglion cells seldom if ever proliferate. Thus, owing either to perplexing artifacts, or to the in- herent complexity of the cortex, its more minute changes seem beyond recognition at present, and when we do de- tect cortical disease processes it is only after they have gone on to some considerable extent beyond the initial stages, and have become rather coarse, extensive, or materially destructive. Since the wonderful revelations of the Golgi methods, one can reasonably enough con- ceive that changes may occur in the cortex which are of the greatest etiological significance, but so subtle that they are entirely hidden from our view. It certainly seems appropriate, therefore, to speak with all this detail about these peculiar difficulties in the way of pathological investigation of the cortex, for if real ad- vances are to be made in the finer pathology of the cor- tex its difficulties of investigation should be appreciated, and if the lesions to be described in these particular cases are to be at all considered as underlying the phenomena of epilepsy, we must approach the problem with all possi- ble caution. I also wish to show that the material placed at my disposal by Professor Starr has such great advan- tages for investigation, both in its structure and prepara- tion, that the difficulties and errors in determining early cortical changes are considerably reduced. From the fact that these minute fragments of the cortex were immediately transferred from the living body to the hardening fluid, the changes in the ganglion cells are es- pecially significant, for the element of artificial change incident to post-mortem alteration or the process of hard- ening larger portions of the cortex, which frequently in- terferes with making positive statements about the minute A CONTRIBUTION TO THE changes in the ganglion cells, is more thoroughly ex- cluded than in the material from an ordinary post-mortem examination. Even allowing for the fact that Midler’s fluid does not preserve the ganglion cells perfectly, the damage to the ganglion cells, presently described, must have existed during life. Microscopic Examination of Case I.—We may now go on with the detailed microscopical examination of the removed portion of the brain in Case 1., and this com- prises a description of: i. A rigid plate of connective tissue acting as a foreign body and pressing against the brain. 2. Changes in the pia mater. 3. Certain lesions of the cortex of the brain, consisting of both changes in the ganglion cells and in the neuroglia. Descriptio7i of the Inwardly Projecting Plate of Connec- tive Tissue Indenting the Surface of the Brain.—The re- moved portion was hardened in Muller’s fluid plus one- sixth its volume of strong alcohol for three weeks. The specimen was very small, measuring about ten by six mil- limetres in diameter, and its central portion furnished about one hundred sections which were cut in series and stained double with hsematoxylon and eosin, and by the picro-acid-fuchsin method. Sections from the centre of the specimen, when recon- structed, show that a tiny plate of very dense, partially calcified connective tissue projected obliquely downward, apparently from the dura mater against the surface of the brain. Here the plate is firmly attached to a minute localized patch of thickened pia mater, and seems directly or indirectly to have pressed on the brain, for the cortex shows an abrupt little pit or depression (see Fig. 1) just beneath the inwardly projecting plate. This cortical de- pression corresponding to the plate is cone-shaped (with the apex projecting inward), and has approximately an altitude of three and three-fourths mm. and a base four to five mm. in diameter. In the individual sections from the centre of the speci- men the plate of connective tissue appears as a very dense, finely lamellated, partially calcified spiculum about three-fourths of a millimetre broad and three millimetres long (see Fig. 1, xx). At its inner extremity the spicu- PATHOLOGY OF TRAUMATIC EPILEPSY. 11 lum has a globular enlargement and the lamellae do not run parallel as in the outer portion, but pass in various directions mostly concentrically arranged about a tiny central nodule or core. The outer end of the spiculum is entirely free in all of the sections, so that it is difficult to determine what the spiculum is a part of, or where it grew from. The inner end of the spiculum is attached in all directions by many diverging fascicles of the thick- ened pia mater. As the sections approach the margin of the specimen Fig. i.—From the Centre of the Removed Portion of the Brain in Case I. The topographical relations of the rigid calcified spiculum of connective tissue, the thickened pia mater, and the depressed region of the cortex, are seen : xx, calci- fic spiculum of connective tissue ; yy, moderately thickened pia mater ; z, anas- tomosing wedge-shaped group of capillaries passing into the cortex from the pia mater, i, 2, and 3, first, second, and third layers of the gray matter ; i, upper portion of the third layer. at one side the plate grows a trifle smaller, but still per- sists to the free edge, so that it seems probable that not all of the plate was removed at the operation. At any rate, it may be said that the removed portion was not large enough to completely surround the plate. From the very dense structure of this connective tissue, and from the fact that the edge of the microtome knife was turned in cut- ting the sections, this plate must have formed a fairly rigid body. 12 A CONTRIBUTION TO THE The Changes in the Pia Mater.—The pia mater, not only at the attached end of the spiculum, but for some little surrounding distance (say three to four millimetres), shows the lesions of chronic meningitis, or productive or hyperplastic inflammation of the pia mater (Fig. i, yy). The pia mater in the region contains an increased amount of connective tissue, which consists of fibro-blasts in dif- ferent stages of development, but most of them show the more mature or final stages. The resultant thickening oi the pia mater, however, is only of a moderate degree, and has not gone on to the extent of obliterating the two layers of the membrane. The inner vascular layer still presents its normal features, although in places (see right- hand portion of the pia mater in Fig. i) the vessels ap- pear to be somewhat diminished in number. The meshes of the inner layer of the pia mater in the depressed region of the cortex are distended and form a network (Fig. i, vv) filled with extravasated red blood- cells. This extravasation of blood, as well as some minute hemorrhages in the gray matter, seem to be of ar- tificial origin, and are very likely referable to the manip- ulation in the removal of the specimen at the operation. The Lesions of the Cortex.—The lesions of the cortex in this case might easily escape detection without the most careful scrutiny and technique. There are hardly any gross changes in the cortex which would attract at- tention with the low power, and it is only with the oil immersion lens that slight changes in the neuroglia cells and scattered damaged ganglion cells become fully ap- parent. These cortical changes are very minute and not at all striking, and yet they are none the less definite and significant. The Ganglion Cells.—The ganglion cells are affected by a series of degenerative changes which in their most advanced stages result in an almost complete dissolution ot the cell, and yet this degeneration is not extensive enough to involve the cells so universally as to interfere with their topographical distribution. Besides this, most of the damaged cells are in the earlier stages of the de- generation, so that they still retain their form and appro- priate position. Thus in reconnoitring the sections with PATHOLOGY OF TRAUMATIC EPILEPSY. 13 the lower powers the ganglion cells do not appear de- ficient in number; they are properly arranged and their several layers are perfectly distinct. The following de- scription applies to all of the ganglion cells excepting the layer of small pyramids. For especial reasons this layer will be dealt with separately later on. It will be convenient to describe the appearance of the nucleus and protoplasm of the degenerated ganglion cells separately, The prevailing form of nuclei shows a dis- tinct peripheral zone, indicating the nuclear membrane : just inside of the nuclear membrane is a narrow clear zone surrounding the chromatic elements of the nucleus, appearing in the form of a skein of finely dotted inter- lacing filaments which show the usual thickened appear- ances at the nodal points and surround unstained inter- stices. The nucleolus is seen in most of these skein-like nuclei, and both the nucleolus and the character of the skein show no variations relating to the different degrees of dissolution of the ganglion cells. In both the early and ultimate stages of the degeneration the form of nu- cleus, as shown in Figs. 2, 3, and 4, remains about the same in all of the cells. This particular form of nucleus in some of the cells is a trifle suggestive of one of the initial stages of kary©ki- nesis, but none of the other stages of mitosis are present, so that this appearance of the nucleus must be regarded as an indication of retrogressive changes. There are no indications of mitosis in any of the ganglion cells, and this agrees with one of the latest papers on the ganglion- cell reproduction by Fiirstner and Knoblauch (Archiv. of Psych., xxiii., 1). Some different appearances of the nucleus are shown in Fig. 2, in the cells v, w, andju The nucleus of the cell w has its chromatic elements resolved into a number (some twenty to twenty-five in optical section) of larger and smaller globules or disks resembling very much the ordinary nucleolus. In the cells y and v the chromatic substance is collected into thickened strands, or large lump-like masses. The protoplasm of the cells shows a series of changes which finally result in an entire disappearance of the cell 14 A CONTRIBUTION TO THE body—for a very complete series of intermediate stages can be observed between the slightly and most com- pletely degenerated cells. The earlier stages of degener- ation consist in larger and smaller solutions of the sub- Fig. 2.—Various Phases of the Earlier Stages of the Degeneration of the Gang- lion Cells. The thin lines enclosing the cells ■?, b, b, vascular layer; d, d, d, and c, layer of clustered neu- roglia islands ; E, E, the compact cortex, largely converted into neuroglia. pursue and by their tendency to unite with each other, they exclude little island-like masses of the cortex. Thus the layer of insular neuroglia masses seems to be formed from the compact brain by a peculiar segregat- ing action of newly formed blood-vessels. It is further to be noted that, following quite universally the layer of insular neuroglia masses, there is always this vascular zone, intimately associated with them, and it lies between the dense masses of connective tissue and the islands. So the capillaries seem to determine the separation of neuroglia masses from the changed solid cortex, and mould them into tiny islands or short cylinders (Fig. n). In Fig. PATHOLOGY OF TRAUMATIC EPILEPSY. 29 11 (taken from the region C, in Fig. 10) a still more de- tailed exposition of this relation of the vessels to the neu- roglia islands is presented. Here the capillaries, one of them collapsed aty,y, are seen surrounding the plugs and islands. But at x, xis shown a stage of subdivision of one of these islands by a solid protoplasmic offshoot of a capillary destined to become hollowed out into a new blood-vessel. Care was taken not to confound this solid protoplasmic process with a collapsed capillary, and it can be observed how it would divide the mass in two nearly equal parts, by uniting with the opposite capillary. Some of these masses of neuroglia persist as isolated little islands even in the midst of dense connective tissue, as shown in Fig. 12. In such instances connective-tis- sue fasciculi have apparently followed the course of the Fig. 11.—The Relation of the Capillaries to the Insular Masses of Neuroglia. y, y, is a collapsed capillary ; x, x, indicates a solid protoplasmic offshoot of a ca- pillary, which is destined to become a new vessel and subdivide one of the insular masses into two portions. capillaries and have grown about the neuroglia islands. Fig. 12 also shows very faithfully the minute structure of these islands. They consist of rather large, glassy neuroglia cells completely enveloped by their own tangled and matted process. In the mass indicated at A, the fila- mentous processes are cut vertically, and at b, two tongue- like processes of neuroglia pass beneath connective-tissue fascicles and fill up two inter-fibrillary spaces. The vessel c has a thickened or hyaline wall. 30 A CONTRIBUTION TO THE Finally, it is to be noted that the cortex, for some little distance surrounding the connective tissue growth, is con- siderably damaged by a degeneration of the ganglion cells and an overgrowth of neuroglia. Remarks.—lt is exceedingly difficult to follow out the connected history of this process in Case 11., without having the opportunity to study the whole of the involved territory of the brain. The rather limited material exam- ined in the removed specimen fails to show the lateral Fig. 12.—The Minute Structure of these Neuroglia Islands and their Persist- ence in the Midst of Dense Connective Tissue. At a the filaments of the neu- roglia cells are cut transversely, and at b two tongue-like extensions of a neuroglia island fill up two inter-fibrillary spaces. border zones of the process, or the relations of the brain membranes, and in general the whole topographical dis- tribution of the lesion. Thus, there is no distinct clew to the origin of the process, or to the formation of the cyst discovered1 at the operation. Still it may be said that this dense mass of connective tissue seems to have grown down rather slowly, apparently from the membranes of the brain, and by disintegrating portions of the brain beneath has contributed to the formation of the cyst, for there is in places distinct evidence of disintegration of the cortex beneath the overgrowth of connective tissue. If in the nature of a scar from the operation, this considerable mass of connective tissue, and its effect on the surrounding PATHOLOGY OF TRAUMATIC EPILEPSY. 31 brain tissue, is very interesting in showing what may hap- pen in the healing process of operations on the brain. It is to the minute changes in the ganglion cells and the neuroglia that attention is more particularly directed, rather than to the grosser changes in these two cases. In reviewing Case 1., these delicate and slightly marked changes on the cortex consist of a little increase in the neuroglia cells, a part of which can be seen on favorable spots only, and a gradual death of the ganglion cells. In describing these two sets of changes in Case 1., it was fre- quently remarked that these minute lesions were of a very early stage of development, and not very striking in the section. This is true as far as our rather coarse percep- tion of them under the microscope is concerned, but when we consider the interference of the mechanism of the cor- tex by these lesions, they are advanced, intense, and very pronounced. Such changes as described in Case 1., if transferred to a comparatively inferiorly-constructed or- gan like the liver, we should hardly associate them with any symptoms ; but taking place in the highly-specialized brain cortex, and in the well-known motor portion, they become invested with a great importance in regard to the production of symptoms. It does not at all convey the full import of the meaning of these minute cortical changes detailed in Case 1., if we finish with them, in summing up the lesions, by simply saying that there is a slight increase of neuroglia and a degeneration of scattered ganglion cells. It is only after the reader conceives of the many initiations or modifica- tions of those molecular changes or nervous impulses which sweep to and fro in the tangled cortical network of cell dendrites and terminal fibre arborizations, that he appreci- ates more vividly the true meaning of these delicate changes which appear so slightly marked under the micro- scope. When we compare the second with the first case, we find the same sort of minute changes in the ganglion cells and neuroglia, in the convolutions around the lateral margins of the mass of connective tissue, and in addition, changes in the white matter of such convolutions. There is then a precisely similar condition of portions of the motor convolutions in both of these cases, consisting in minute 32 PATHOLOGY OF TRAUMATIC EPILEPSY. changes in the neuroglia and ganglion cells. In the first case the condition is referable apparently to the foreign body and the trauma. In the second case the influence of the connective-tissue mass seems to have in- duced the condition in the neighboring convolutions. It seems to me that this gradual death of the ganglion cells in the motor zone, especially the very large motor cells, together with the growth of neuroglia, in both the gray and white matter explains the symptoms of epilepsy very well, and much better than some of the other de- scribed changes, as for instance the lesions of the cornu ammonis, which certainly have very little causal relation to the symptoms as far as localization is concerned. Of course these are traumatic cases, but this ought not to preclude regarding this disease process in the cortex in a significant relationship to the symptoms of epilepsy. For trauma might well enough be only one of several con- ditions which produce this same sort of disease-process in the motor cortex, resembling somewhat a very com- mon form of inflammation of the internal organs exem- plified by chronic diffuse nephritis, which arises a variety of conditions for the most part not understood. Nevertheless, while it is very tempting to see in these lesions a basis for the motor phenomena of epilepsy, there are not facts enough in these two isolated traumatic cases to say anything positive about this relation. Still I ven- ture to think that such changes as described in these cases are very suggestive lines along which to work out the true lesions in idiopathic epilepsy. Idiopathic epi- lepsy certainly seems to be a disease of the motor zones, and it behaves like an organic disease with definite lesions behind it. A question which presents itself in connection with Case 11. is, how often do the operations on the brain cortex leave an irritating scar which tends to produce lesions in the surrounding brain tissue like those de- scribed in the convolution A, Fig. 9 ? If trephining the motor cortex for traumatic epilepsy leaves a cicatrix ir- ritating the neighboring convolutions and changing them like this convolution A, the epileptic symptoms ought to return in a certain length of time after the operation.