tation can be put upon this experiment-we have no ground for supposing even in the case of human epilepsy that a pure tonic seizure can arise and run its course without the participation of an excitation of the cortex.

In sum, it may be said that the tonic component of the epileptic seizure originates and is maintained under the direction of excitation of the cortex, but that it may also be maintained by means of a subsequently independent excitation of sub-cortical areas.

To consider a little further the clonic component of the convulsion; if it be assumed that the cerebral cortex manifests a refractory phase of not inconsiderable duration, then the series of clonic twitches such as are induced, for example, by a caloric stimulus would correspond fully to what we observe when the heart muscle is stimulated with a galvanic current. The cortical stimulus gives rise to such a momentary muscular contraction, after which the cortex is refractory for about half a second; at the end of this refractory period the stimulus acts in the manner of a new stimulus suddenly applied, and the cycle is repeated. The infra-cortical structures, however, may have no refractory phase or only a very brief one. If these are brought into an independent state of excitation, they will transmit the latter to the nerves and muscles in the form of a continuous excitation of extremely rapid rhythm, so that the muscles fall into a state of tonus-the "tetanus" of muscle physiology. As a matter of fact, Broca and Richet have been able to demonstrate a refractory phase of appreciable duration by electric stimulation of the cortex. Zwaardemaker has shown the same to be true of the swallowing and winking reflexes; in both these instances the refractory phase was particularly long, one to three seconds in duration, while in the experiments of Broca and Richet it was of 0.1 to 0.7 seconds' duration, according to the temperature. Further investigation, with some inquiry into the bearing of the strength of the stimulus, would be necessary to obtain further insight into the relationship of these phenomena to clonus."

How is the excitation transmitted from the primary site of stimulation in the cortex? Ewald obviated the spread of the stimulus itself by implanting in the cortex small thin-walled glass tubes which projected slightly above the surface; he thus indicated the

5. That a phase of increased excitability follows the refractory phase has been shown by Isayama. In his recently published book on epilepsy, Muskens has treated the question of the refractory phase in some detail.

site of primary stimulation, not only with reference to the transmission of the excitation via the association fibres, but also to the spread of the stimulus by conduction of the current. Under these circumstances spread of the convulsive excitation no longer occurred. That the excitation is disseminated in the homolateral cortex itself is likewise suggested by the demonstration of Ziehen and by Bubnoff and Heidenhain that the further spread of the convulsion could be interrupted if during its course the primary site of excitation were quickly extirpated; furthermore, no clonic movements made their appearance in the muscle groups of which the corresponding cortex had been removed.

As to the passage of the excitation to the opposite half of the brain, Lewandowsky declared that the corpus callosum acts as the medium of transmission, and Horsley's experiments on dogs suggest the importance of this structure in effecting this transmission. Horsley found that after cutting through the corpus callosum and the commissures the effect of the cortical stimulus in the form of tonus followed by clonus appeared in full only in the contralateral extremities, while the effect upon the side of the stimulus consisted merely in a condition of feeble tonus or in none at all. Karplus, however, was unable to verify this observation in either the dog or the monkey; he demonstrated that after cutting through the corpus callosum, stimulation of the right half of the cortex gave rise to clonic movements on the part of the muscles of the left side of the body, while the extremities on the right side might also become involved, although not with the same intensity as upon the left side. If in addition the right cortex was removed and the left then stimulated, the extremities controlled by the right cortex still participated in the tonic phase of the convulsion but no longer in the clonic. Hence it is apparent that even after cutting through the corpus callosum the opposite half of the cortex takes part in the excitation. Spiegel obtained a similar result and showed besides that the generalization of the convulsion was not prevented by division of the commissures. We see, therefore, that the corpus callosum does not take part in the transmission of the epileptic excitement-or, more accurately expressed, we are unable to demonstrate its participation. Much remains obscure with reference to the pathways by which the excitation aroused by the original stimulation of the cortex is transmitted to the sub-cortical region, the structures in the

latter which are involved in the transmission of the excitation to the opposite hemisphere, and the pathways by which the excitation then reaches the cortex of that hemisphere.

As to the descending pathway along which the excitation is conducted, the pyramidal tracts would naturally occur to one as being the most important; but these do not appear to have the preponderant role formerly attributed to them. Hering showed that clonic spasms could occur in a dog when the pyramidal tracts had been disconnected; the converse was also true, and the pyramidal tracts mediated the clonic spasm when the remaining tracts were out of circuit. Economo and Karplus showed that all the usual consequences of electrical stimulation of the cortex took place in the cat when the pedunculi cerebri were cut through. The pyramidal tracts play a more important part in the monkey, however, for, according to Hering, clonic movements of the extremities no longer occur when these have been divided, though still observable in the head, eyes and face.

Regarding the point at which the excitation passes to the opposite side, Karplus has shown that it is not located in the spinal cord. Spiegel has furthermore shown that none of the transverse tracts anterior to the rhombencephalon are required for this. How the excitation reaches the rhombencephalon from the cortex is still unknown. The frontal and temporal cortico-pontine tracts are not involved.

One important question remains to be considered: In the production of the tonic component of the convulsion, is the same mechanism in question as produces tonic excitation in various other conditions? Decerebrate rigidity comes to mind as an example of the latter-likewise insulin convulsions, the convulsions associated with hyperventilation and asphyxia, and parathyreoid tetany. Without discussing this complicated question at any length, it may be said that it is not probable that a mechanism is operative in the tonic seizures of epilepsy which is dissimilar from that present in other conditions of muscular hypertonia, nor is it very likely that this mechanism is widely different from that of the maintenance of normal tonus. But here we must confine ourselves to the question of the relation of the tonic spasm in epilepsy to decerebrate rigidity. While no entirely satisfactory answer has as yet been forthcoming to the question whether decerebrate rigidity is a phenomenon

of stimulation or of inhibition (the experiments of Economo and Karplus, Bazett and Penfield, and Magnus and Rademaker, who produced decerebrate rigidity of considerable duration, suggest the latter explanation), the locus of the phenomenon, at least, has apparently been established. Rademaker has shown that tonus remains unimpaired when section is made directly anterior to the red nucleus, but that decerebrate rigidity makes its appearance when it is made immediately caudal to it. Section of the rubrospinal tracts at their decussation likewise produces decerebrate rigidity. There can be no doubt but that the red nucleus plays an important part in this phenomenon. If we assume with Rademaker that this is also the case in man, the question whether stimulation or a cutting out of circuit is involved is not pertinent to the inquiry as to the relation to decerebrate rigidity of the tonic component of the epileptic seizure, for epileptic tonus is certainly a phenomenon of "stimulation" in the usual sense of the word. Nevertheless, further investigation is required before we can approximate a solution of the complex question of the relation of epileptic rigidity to other rigidity states. The situation is made more difficult by the fact that, apart from the red nucleus, we know too little regarding which structures in the deeper portions of the central nervous system are involved in the production of tonic manifestations. The same uncertainty, indeed, confronts us here as did regarding the question of the point at which the convulsive excitation passes from the site of stimulation in the brain to the opposite hemisphere, of the manner in which the excitation arrives at this point, and of how it is then transmitted to the cortex of the opposite hemisphere. The locus in the nervous system of the excitation which underlies the phenomenon of rigidity is presumably the same in epilepsy, decerebrate rigidity and tetany, and likewise in other conditions (asphyxial and hypoglycemic convulsions, for example) with which increased tonus is associated. According to Spiegel and to Kleitmann and Magnus, this locus is to be sought in the rhombencephalon.

THE REGULATION OF CHLORIDE-BROMIDE INTAKE

IN EPILEPSY*

BY. DR. J. NOTKIN,

SENIOR ASSISTANT PHYSICIAN, MANHATTAN STATE HOSPITAL, NEW YORK CITY; ASSOCIATE IN NEUROLOGY, NEW YORK POST-GRADUATE MEDICAL SCHOOL AND HOSPITAL

As the cause of epilepsy is still undetermined, therapy necessarily remains symptomatic and must limit itself to the control of the convulsive seizures. A number of drugs have been advocated for this purpose, among others, at first bromides, in later years, Luminal. Simplicity of technique with the latter form of medication won great favor for this drug. However, complete cessation of seizures with Luminal has been reported only in isolated cases, and in numerous others it did not reduce their number to any great extent. A more efficient drug was sought; but in the absence of one, as frequently happens, recourse was again had to the older remedy, and renewed attempts were made to improve the technique of bromide medication. After the experimental work of the Zurich school became known, bromine came again into the limelight. The early difficulties with bromine were largely due to the fact that at first the drug was used rather indiscriminately. Even today, there still exists much confusion as to the physio-chemical processes set up by this medication. In regard to this intricate problem we shall not go into the controversy among different workers. It may merely be mentioned here that two theories have been advanced, one-the theory of hypo-chloridation as a result of bromide medication, the other-the theory of the specific action of bromides on the central nervous system. As to the theory of hypo-chloridation, our experiments do not indicate that this is an important factor in every case. Out of seven patients who were put on a high table salt diet, (from 20.0 to 30.0 grams a day) 3 had an increased number of seizures, while 4 showed no response at all. The fact that over 50 per cent did not respond to the hyper-chloridation would point to a different pathology in these cases and may perhaps give us an idea of a difference in etiological factors in the epilepsies.

Bernoulli and Ulrich, who in recent years have done considerable work on this problem, state that neither hypo-chloridation alone

* Read at the annual meeting of the National Association for the Study of Epilepsy at Cincinnati, May 31, 1927.