Aster periods of disinterest, neurosurgeons' attention to the corpus callosum has been reawakened for a variety of reasons. For centuries, the large size, central location, and widespread connections of the corpus callosum stimulated investigations, which were motivated as much by scientific curiosity as by therapeutic considerations. Callosal physiology has more recently been important to surgeons concerned primarily with other structures, including those neighboring the third ventricle, which can be approached through the corpus callosum. But the principal motivation has been the role of the corpus callosum in the generation of seizures. At the end of the 19th century and on two other separate occasions in the 20th century, callosotomy was considered as a treatment for seizure disorders. Before 1900, less obviously around 1940, and most clearly in the 1960s, these therapeutic considerations were stimulated by animal experimentation. One theme of this chapter is the reciprocal interaction of surgical therapy and laboratory experimentation: in particular, the most recent therapeutic use of callosotomy has been accompanied by widespread physiological and psychological interest in this conspicuous brain structure.

In this chapter we consider first a brief history of studies of the corpus callosum. Then follows a chronological account of interest in callosotomy as a treatment for epilepsy. I will conclude with a few personal observations.

Studies of Callosal Function

Studies of the corpus callosum were first undertaken by the Humoral Anatomists. These were the writers of antiquity whose concepts of brain function emphasized the contents of the brain cavities and the flow of various fluids such as air, phlegm, cerebrospinal fluid, and blood. For them, the corpus callosum seemed largely a supporting structure. This view persisted for a millennium. Even that original Renaissance genius, Andreas Vesalius (1514-1564), believed that the corpus callosum served mainly as a mechanical support, maintaining the integrity of the various cavities. In 1543, he wrote:

There is a part [whose] external surface is gleaming white and harder than the substance on the remaining surface of the brain. It was for this reason that the ancient Greeks called this part "tyloeides" ["callosus" in Latin] and, following their example, in my discourse I have always referred to this part as the corpus callosum. If you look at the right and left brain ... and also if you compare the front and rear, the corpus callosum is observed to be in the middle of the brain; ... Indeed, it relates the right side of the cerebrum to the left; then it produces and supports the septum of the right and left ventricles; finally, through that septum it supports and props the [fornix] so that it iyian, not collapse and, to the great detriment of all the functions of the cerebrum, crush the cavity common to the two [lateral] ventricles of the cerebrum.'3(p597)

In the 17th century, the "traffic anatomists" took a major step forward. It was at about the time of Thomas Willis (1621-1675)that anatomists began thinking more in terms of a traffic or communication between the more solid parts of the brain.46 This view became quite explicit in the statement of Felix Vicq d'Azyr (1748-1794) who wrote in 1784:

"It seems to me that the commissures are intended to establish sympathetic communications between different parts of the brain, just as the nerves do between different organs and the brain itself . . ."13(p592)

For over two centuries, beliefs about callosal function consisted almost solely of inferences from its central location, widespread connections, and large size (larger than all of those descending and ascending tracts, taken together, that connect the cerebrum with the outside world). Among others, Willis, François de la Peyronie (1678-1747), and Giovanni Lancisi (1654-1720) thought the corpus callosum a likely candidate for "the seat of the soul," or they used some other expression intended to cover that highest or ultimate liaison which brings coherent, vital unity to a complex assemblage. 13

The observations of the early anatomists have often been supported by subsequent anatomical observations, including the large number of callosal fibers (at least 200 million of them). Because the callosal fibers interconnect so much of the cerebral cortex, especially that cortex considered associative, it has often been suggested that they serve some of the "highest," most educable, and characteristically human functions of the cerebrum.

Inference of function from observable structure is time-honored and productive; however, such inference has its limitations. The physiological evidence has only partially sustained anatomical inference. We now know from various observations (notably the split brain) that the corpus callosum is indeed an important integrative structure; we also know that it is neither sufficient nor indispensable, providing only one of a number of integrative mechanisms.

That the corpus callosum is not the exclusive "seat of the soul" is evident from the apparent normality in social situations of patients who have had complete callosotomies.4,5 That it is an important integrating mechanism is clear from the peculiarities of such patients. These include, among other things, a unilateral tactile anomia, a left hemialexia. and a unilateral apraxzia. That is, for the right-hander with complete callosotomy, there is an inability to name aloud objects felt with the left hand, an inability to read aloud written material presented solely to the left half-field of vision, and an inability to execute with the left hand actions verbalIy named or described by the examiner. The apraxia usually recedes in a few months, whereas the hemialexia and unilateral anomia persist for years. Such deficits are now readily demonstrable in individuals who have had surgical section of the corpus callosum. But these deficits were first recognized in patients with vascular disease that caused very complex and evolving syndromes.

In the closing decades of the 19th century (or more broadly construed, in the period between the American Civil War and World War 1) there emerged that group of neurologists whose discoveries and formulations are still at the core of current clinical knowledge. Among them were Carl Wernicke(1848-1905), Hugo Liepmann(1863-1925), J. Jules Dejerine (1849-1917), and Kurt Goldstein (1878-1965), who interpreted various neurological symptoms as resulting from disconnection, including interruption of information flow through the corpus callosum.

The concept of apraxia was developed by Liepmann expressly to describe a patient who could carry out commands with one of his hands but not with the other. In 1908, Liepmannand Maas 26 described a right-handed patient whose callosal lesion caused a left apraxia as well as a left-handed agraphia (an inability to write) in the absence of aphasia. These disabilities have subsequently been observed many times. Unilateral apraxia and unilateral agraphia are not always present, and they may subside when a stroke victim progressively recovers, but they remain among the cardinal signs of callosal interruption.

Liepmann considered the corpus callosum instrumental in most left-hand responses to verbal command: the verbal instruction was comprehended only by the left hemisphere, and the left hand followed instructions delivered not by a directly descending pathway (which we now call "ipsilateral control") but by a route involvig callosal interhemispheric transfer from left to right and then by right hemisphere control of the left hand (what we now call "contralateral control").

Necessarily then, callosal interruption would result in an inability to follow verbal commands with the left hand, although there would be no loss of comprehension (as expected from a left hemisphere lesion) and no weakness or incoordination of the left hand (as could result from a right hemisphere lesion). This view was largely ignored or rejected (particularly in the English-speaking countries) for nearly half a century.

Norman Geschwind (1926-1984) suggested that there was a widespread revulsion against attempts to link brain to behavior, associated with the rise of psychoanalysis (personal communication). He had another sociological explanation:

Henry Head had been shrewd enough to point out that much of the great German growth of neurology had been related to their victory in the Franco-Prussian war. He was not shrewd enough to apply this valuable historical lesson to his own time and to realize that perhaps the decline of the vigor and influence of German neurology was strongly related to the defeat of Germany in World War I and the shift of the center of gravity of intellectual life to the English-speaking world, rather than necessarily to any defects in the ideas of German scholars. 19 As Harrington put it, 2l ways of thinking about the brain (i.e., laterality and duality) which seem natural enough now had "vanished from the working world view" for nearly 50 years. She has made available in scholarly detail the popularity of these ideas (laterality and duality) in the 19th century, and their eventual re-emergence in the 1960s. Chapter 9 of Harrington's book2l is devoted to the causes of this long eclipse. She was particularly critical of Henry Head (1861-1940), whose highly selective reference to John Hughlings Jackson, she wrote, "borders on intellectual dishonesty."

Besides the sociological aspects, other factors were involved. There was widespread reluctance to consider callosal disconnection as the efficient cause of deficits such as unilateral apraxia or hemialexia occurring in patients with lesions (e.g., tumors or infarctions) involving the corpus callosum. This reluctance developed in large part because surgical interruption of the corpus callosum had not been found to cause the same deficits.

Walter Dandy (1886-1946) went so far as to say in 1936: "The corpus callosum is sectioned longitudinally... no symptoms follow its division. This simple experiment puts an end to all of the exravagant hypotheses on the functions of the corpus callosum."14 Even more persuasive were the negative tests performed by Andrew J. Akelaitis (1904-1955)1 on patients who had callosal section. By the end of the 1950s, Fessard summarized the view that was then generally accepted: ". . . there is a great deal of data showing [that] section of important associative white tracts such as the corpus callosum does not seem to affect mental performances. Other similar observations in man or animals are now accumulated in great number and variety. "17

We now realize that most of the negative findings resulted from two sources:

1. When surgical section of the commissures is incomplete, a remarkable capacity for maintaining cross-communication between the hemispheres may be retained with quite small cmmissural remnants, particularly when the part remaining is at the posterior end of the corpus callosum (in other words, the splenium).

2. Negative findings often result from the use of inappropriate or insensitive testing techniques. What one finds depends on what one looks for; although Dandy" said that callosal section produces no observable deficits, among his own patients was one reported by Trescher and Ford to have hemialexia.

Essential to the resurrection of the callosal disconnection view was the ability to observe repeatedly and appropriately under controlled, prospective circumstances the results of callosotomy in humans. This was facilitated by the use of complete callosotomy as a treatment for epilepsy, which, in turn, had been made possible by cat and monkey experiments beginning in the 1950s. 3 Forty years of experimentation with laboratory animals and 30 years of experience with callosotomized humans have by now firmly established the principal features of callosal section and facilitated the more precise interpretation of deficits following naturally occurring lesions.5

The impression is gained from this small number of observations that the type of case in which section of commissural fibers in the corpus callosum is most favorable is the one in which a large cortical or subcortical scar exists, as in cases I and 10.

The Corpus Callosum and Epilepsy

The famous experiments of Gustav Fritsch (1838-1891) and Eduard Hitzig (1838-1907) in 1870 showed that there was a limited region of the cerebral cortex (the "motor cortex"), electrical stimulation of which resulted in movements of the contralateral limbs."," In two of their dogs, removal of the electrically excitable cortex resulted in subsequent incoordination of the contralateral limbs. In another two dogs, tetanization caused "epileptic attacks" beginning first in the contralateral limbs and then becoming generalized. The generalization of the "epileptic attacks" could be interpreted, in retrospect, as due to transmission across the corpus callosum.

A next step was taken by Bubnoff and Heidenhain in 1881. They stimulated the white matter exposed by ablation of the motor cortex. This stimulation could produce convulsive movements in the unparalyzed ipsilateral limbs. They concluded that excitation had spread across the corpus callosum to involve the motor cortex of the opposite hemisphere. A few years later, in 1886, Sir Victor Horsley24 (1857-1916) lectured on the effectiveness of cutting the corpus callosum to prevent the spread of seizures. He recognized (as had Bubnoff and Heidenhain) that subcortical circuits could maintain convulsive activity once it began, but emphasized the role of cerebral cortex in the initiation of convulsions.

 In succeeding decades, the possible role of the corpus callosum in seizure spread was studied in experimental animals by many other investigators. When Spiegel" reviewed the physiology of epilepsy in 1931, he emphasized the common finding that generalized convulsions could occur after the corpus callosum connections had been severed. His own experiments included section of all crossing fibers down to the rhombencephalon and he stated that "even after this operation, we could observe that general clonic convulsions developed following one-sided cortical stimulation." He also described experiments with a sagittal section of the rhombencephalon in the midline, again with generalized convulsions being possible from stimulation of one hemisphere. Evidently, generalization can occur through fibers that cross the midline at several levels. Many years before, John Hughlings Jackson (1835-1911) had asserted that whatever its behavioral manifestations, seizure activity was characterized by an excessive discharge (we now call it "hypersynchronous and self-maintaining") of nerve cells. In 1878, he wrote: "A convulsion is but a symptom, and implies only that there is an occasional, an excessive, and a disorderly discharge of nerve tissue on muscles."24 (p8) Elsewhere he wrote:

Scientifically, I should consider epilepsies on the hypothesis that the paroxysm of each is dependent on a sudden temporary excessive discharge of some highly unstable region of the cerebral cortex. There is, in other words, in each epilepsy a "discharging lesion" [which] leads to secondary discharge of healthy cells in other centers . . .. 23(p276)

This assertion met with considerable resistance at the time but was ultimately confirmed by electroencephalographic (EEG) studies. These included animal experiments by Moruzzi,28 among others. Moruzzi observed that epileptiform EEG activity induced by electrical stimulation in one hemisphere promptly appeared in the other with a latency consistent with the conduction speed of callosal fibers, and this spread could be prevented by callosotomy. However, this finding did not negate the possibility of spread via other routes.

The multiplicity of routes for seizure spread rendered unattractive the idea of cutting the corpus callosum for treatment of epilepsy, particularly in view of the technical difficulties which this would involve. The feasibility of cutting the corpus callosum was subsequently emphasized in 1936 by Dandy, who used this approach to reach midline tumors. But neither he nor others of the time reccommended callosotomy for epilepsy.

The first callosotomies for epilepsy were reported in 1940 by Van Wagenen and Herren; 42 they operated on their first case on February 6, 1939. William P. Van Wagenen (1897-1961) was at that time Chief of Neurosurgery at the University of Rochester, Strong Memorial Hospital. The rationale given for this procedure was their observation of two cases in which a callosal tumor had lessened seizure frequency and two cases of vascular insult that stopped the seizures altogether. Their paper contained no references, hence it is not clear if the authors knew of Spiegel's 1931 report 39 or the material that he summarized. It seems less likely that they were aware of Mortuzzi's investigations in the mid-1930s. Nor is it clear if they were aware of the work of Erickson. In 1940, Theodore Erickson (1906-1986)16 had performed experiments on monkeys showing a role of the corpus callosum in seizure spread, and his report was published in the preceding volume of the same journal. One might assume that Erickson's animal experiments at the Montreal Neurological Institute were known to Van Wagenen, since the neuroscience community was quite small in those days. Moreover, both attended meetings of the American Neurologic Association. Robert Joynt,25 who arrived at the University of Rochester as Chairman of Neurology in 1966, hasemphasized that the Van Wagenen series was undertaken "solely on clinical observations [which were personally made] by the two authors." Joynt focused special attention on the summary of the Van Wagenen-Herren paper. Parts of the summary are worth repeating here because their conclu;sions have remained largely correct after more than 50 years.

Van Wagenen and Herren summarized their first 10 experiences with callosotomy as follows:

The impression is gained from this small number of observations that the type of case in which section of commissural fibers in the corpus callosum is most favorable is the one in which a large cortical or subcortical scar exists, as in cases I and 10.

Whether section of various commissural pathways to prevent the spread of an epileptic wave is indicated for patients having multiple irritable fociis a matter for future study ... the observation on patients having jacksonian seizures on the right side after section of the corpus callosum on one occasion and on the left side on another suggests that there are at least bilateral foci from which seizures may originate (case 7).

The inhibitory effect of the cortex of one hemisphere on the activity of the other must also be considered seriously ... it may be that in certain instances the cortex of one hemisphere may inhibit abnormal activity of an abnormal zone and that section of commissural pathways is contraindicated.

Section of the commissural pathways contained in the corpus callosum may be carried out without any untoward effect on the patient. Such a section may serve to limit the spread of an epileptic wave to the opposite hemisphere. When such limitation occurs, the patients do not seem to lose consciousness or have generalized convulsions. 42(pp758-759)

Although Van Wagenen and Herren seemed pleased with their results, no one else took up the operation. This may have been in part because of the onset and continuation for 5 years of World War 11, or longer term outcomes may have been unfavorable and generally known although unpublished (see the section on personal recollections). An important factor was the growing conviction, reaching a peak in the 1950s, that the reticular formation and its rostral targets in the thalami are of particular importance for seizure spread. As Penfield and jasper stated in one of the great classics of epileptology: It seems reasonable to assume, therefore, that generalization of the motor seizure does not take place by spread of excitation through cortical circuits. It must spread through the more closely interrelated neuronal network of the higher brain stem, in a centrencephalic system with symmetrical functional relationships to both sides of the body."'

The irrelevance of callosal transmission for generalization of unilateral seizures was also suggested by the occasional recurrence of generalized convulsions in humans with hemispherectomy.45 Indeed, in a few experiments generalized convulsions could be produced by Van Harreveld in dogs subsequent to bilateral decortication .41

In spite of the foregoing, a number of other considerations discussed below suggested the possibility of improvement of otherwise untreatable seizure disorders in carefully chosen cases. Briefly, the reintroduction in 1962 of callosotomy together with anterior and hippocampal commissurotomy as a treatment for medically intractable epilepsy was stimulated and indeed made possible by animal experimentation involving similar procedures in cats and monkeys (e.g., Sperry and Myers 29,.,6) .

Because of the remarkable improvement of two patients treated by Bogen et al6,9 we continued to offer the operation. A few years later we briefly reported our rewarding results in nine of 10 cases, each having at least 2 years followup, concluding: "It thus appears that the combination of cerebral commissurotomy plus postoperative medication has limited propagation of seizure activity from a cortical focus.8

The improved status of these patients made possible their participation in a long and still continuing series of neuropsychological investigations.5,38, 48,49 These investigations, which contributed to Roger Sperry's Nobel prize in 1981, supported two generalizations: that there was incomplete but substantial hemispheric independence, and complementary hemispheric specialization. In a classic paper, Sperry wrote:

... Although some authorities have been reluctant to credit the disconnected minor hemisphere even with being conscious, it is our own interpretation based on a large number and variety of nonverbal tests, that the minor hemisphere is indeed a conscious system in its own right, perceiving, thinking, remembering, reasoning, willing, and emoting, all at a characteristically human level, and that both the left and the right hemisphere may be conscious simultaneously in different, even in mutually conflicting, mental experiences that run along in parallel.

Though predominantly mute and generally, inferior in all performances involving language or linguistic or mathematical reasoning, the minor hemisphere is nevertheless clearly the superior cerebral member for certain types of tasks.... Largely they involve the apprehension and processing of spatial patterns, relations, and transformations. They seem to be holistic and unitary rather than analytic and fragmentary, and orientational more than focal, and to involve concrete perceptual insight rather than abstract, symbolic, sequential reasoning. However, it yet remains for someone to translate in a meaningful [i.e., physiological] way the essential right-left characteristics. . . .37(p11)

1) The data, as well as what they implied, were sufficiently dramatic to attract increasing media attention. Much of this was hastily written and often sensationalized. According to the New Yorker magazine of November 8, 1976 (p 36), "The corpus callosum is an inch long An article in the New York Times Magazine on September 9, 1973, included an artist's drawing of a split-brain patient wielding a hatchet in the left hand which was being restrained by the right hand. Less esteemed media outlets were worse, and there were innumerable cartoons. The media pushed the popularity of the "right brain/left brain" story to fad proportions, reaching an almost frenzied peak by 1980. This led not only to simplistic degradation, probably inevitable with popularization, but also to exploitation. Commercially motivated entrepreneurs promised to educate people's right hemispheres in short order, sometimes even overnight, ignoring the lengthy, arduous training necessary for mature competence.

This was followed by a reaction or backlash, much of which involved the debunking of extravagant claims." Some of it, however, was more revisionist; that is, some writers challenged the basic observations. A notable example is the recent explicit rejection of hemispheric specialization by Efron.15 I have offered elsewhere 5 some tentative evaluations of these events, which provide an example of how 20th century neurosurgery has influenced both academic and popular psychology.

Our reports of therapeutic success with callosotomy were followed by a few others (e.g.,Luessenhop et al27) . However, wider acceptance of this procedure was made possible only by the sustained effort during the decade of the 1970s of Donald H. Wilson (19271982) and coworkers, particularly Alexander Reeves, at Dartmouth Medical 17School. By 1982, at the Dartmouth conference organized by Wilson and Reeves, the use of callosotomy for epilepsy was reported from six more clinical centers. There had also been a burgeoning of experimental studies, many of them presented at that conference. Both the clinical and experimental contributions were subsequently included in a book edited by Reeves. 31 During the 1980s, callosotomy (either complete or more often partial) became an established procedure. In the words of Spencer et al:

Over the past decade, corpus callosum section has become a widely accepted, relatively safe, clearly needed, broadly practiced, and continually evolving addition to the medical treatment of certain types of severe and uncontrolled seizures in certain types patients who are not candidates for resective procedures.35 Over a span of about 30 years, during which callosotomy came to be more favorably viewed as a treatment for epilepsy, increased familiarity with both the anatomy and the physiology of the corpus callosum has encouraged the use of partial callosotomy as an approach to the third ventricle and other midline structures.2

Some Personal Recollections

In 1959, I wrote an essay entitled, "A Rationale for Splitting the Brain in Cases of Epilepsy." I took it to Roger Sperry, whom I had known for several years at the California Institute of Technology and who was, I felt, the neuroscientist of our time. He offered two suggestions: "Change the title" and "Look up those papers by Akelaitis." My subsequent review of the Van Wagenen-Akelaitis series is detailed elsewhere,32 and can be briefly summarized as follows:

1) The highest proportion of poor results was in patients having only a partial section.

2) Of those surviving a nearly complete operation and followed for up to 2 years, two-thirds were improved, including one-third who were free of generalized convulsions.

3) Unilateral seizures were quite common after callosal section, either on one side or alternating on both sides.

Meanwhile, there was the question of who would be a suitable candidate for such a procedure; it turned out to be William Jenkins.' I first met Bill Jenkins in the summer of 1960when he was brought to the emergency room in status epilepticus; I was the neurology resident then on call. The heterogeneity as well as the intractability and severity of his multicentric seizure disorder became clearer to me over the next months. Both in the clinic and in the hospital I witnessed psychomotor spells, sudden tonic falls, and unilateral jerking, as well as generalized convulsions. In late1960, I wrote to Maitland Baldwin, then Chief of Neurosurgery at the National Institutes of Health (NIH) in Bethesda, Maryland. A few months later, Bill was admitted to the NIH epilepsy service where he spent 6 weeks. He was sent home in the spring of 1961, having been informed that there was no treatment, standard or innovative, available for his problem.

Bill and his wife Fern were then told of Van Wagenen's results, mainly with partial sections of the cerebral commissures. I suggested that a complete section might help. Their enthusiasm encouraged me to approach Professor Philip J. Vogel (b. 1906),because of his experience with removal of callosal arteriovenous malformations. He was Chief of Neurosurgery in the Loma Linda University Medical School and Director of Neurosurgery at the White Memorial Hospital, at that time the teaching hospital for Loma Linda University. He suggested that we practice a half-dozen times in the morgue. By the end of the summer (during which I was again on the neurosurgery service) the procedure seemed reasonably in hand. There was further delay during which Bill underwent testing in Sperry's laboratory, mainly with Michael Gazzaniga who had arrived in September as a beginning graduate student in psychobiology. During this delay we also had an opportunity to keep a reasonably complete record of Bill's many seizures.3

It was during this period of preoperative testing that Bill said, "You know, even if it doesn't help my seizures, if you learn something it will be more worthwhile than anything I've been able to do for years." He was operated on in February 1962. It seems to me in retrospect that, if there had been a research committee at our hospital whose multimember approval was required, the procedure would never have been done. At that time, a chief of service could make such a decision alone, which I expect was similar to the situation at the University of Rochester in the late 1930s.

From the start, our procedure included not only complete callosotomy (requiring two skull openings) but also section of the anterior commissure, accessed in most cases by entering the third ventricle through its roof. We chose to perform as complete a section as possible for two reasons:

1) monkeys undergoing this procedure were without neurological disability and participated well in demanding psychological testing, 16 and
2) if a complete neocommissural section failed in this ideal case (an intelligent, personable individual with supportive family whose multicentric seizure disorder could hardly have been much worse) then we would be through. Fortunately, it succeeded.

The completeness of our procedures as subsequently confirmed by magnetic resonance imaging was attained without the use of the operating microscope (which I first used in 1970), the good light that the scope provides, the bipolar cautery, osmotic diuretics, modern neuroanesthesia, and a variety of instruments only subsequently available. This is a tribute to Vogel's operative skills, including his sense of tolerable retraction and his remarkable vision at the usual operating distances (he never did take up the microscope). And how impressive it is that Van Wagenen worked under even less auspicious conditions!

Our next major step was to do a callosotomy (and anterior commissurotomy) which spared the splenium, whose section we believed by then to be the main source of disconnection symptoms. Throughout the 1960s, Vogel and I had been approaching lesions in or near the third ventricle via lengthy incisions in the middle of the corpus callosum; these patients did not show the disconnection effects of the complete section. I became increasingly confident of this conclusion, having by then considerable practice in detecting the disconnection effects by bedside examination.

These clinical findings had been stimulated by and gave increasing support to the view that the negative results of Akelaitis were not solely attributable to his lack of appropriate testing techniques. His negative results seemed also ascribable in part to the incompleteness of many of Van Wagenen's callosotomies, often described as "nearly complete" or as involving all but the most posterior end of the corpus callosum. By 1968, these considerations led to the expectation that section sparing the splenium could avoid most of the disconnection syndrome while at the same time ameliorating seizures having a rostral origin. Specifically relevant were complex partial seizures involving both anteromedial temporal regions, without generalization to the entire cerebrum when the patient was adequately medicated.

In 1968 and 1969, we operated on two patients whose seizures caused life-threatening psychomotor behavior and whose bitemporal foci appeared to be independent. Their seizure disorders were markedly improved (one subsequently obtained a steady job for the first time) and they had no discernible disconnection symptoms.20 In the words of Wada, this report ". . . revitalized our interest in re-examining brain bisection as a possible new treatment modality. . . . "44

By now, sparing both splenium and anterior commissure has become commonplace, particularly because a section restricted to the anterior two-thirds to three-fourths of the corpus callosum can alleviate drop attacks, and drop attacks are in the opinion of many the prime indication for callosotomy. That drop attacks could be eliminated by callosotomy never occurred to us, even by 1974 when we summarized our criteria for operation." This was in spite of the fact that the commissurotomy eliminated Bill Jenkins' drop attacks as well as his generalized convulsions (except for two occasions in 10 years). We were still influenced to some extent by the concept of "centrencephalic seizures," our theoretical views preventing us from recognizing a fact in front of our eyes.

The idea that extremely rapid generalization of seizures required a centrencephalon weighed even more heavily with others than with us, and was probably responsible in part for the disbelief with which our reports were received. In addition, our work was done at a medical school (Loma Linda University) better known in those days for training medical missionaries than for scientific advances.

Not only was the procedure at odds with a wellknown theory, it was worse! Had not this approach already been tried and failed? When I wrote to Frank Smith, then Chief of Neurosurgery at the University of Rochester, asking for as much information as he could provide, his reply was quite short, including that, "Dr. Van Wagenen always was sorry about what he did to those patients" (see also Smith"). For over a decade there were persons in Boston who referred to us as "the West Coast butchers." Without the excellent work of Wilson and Reeves, it is quite likely that our efforts (as well as of others, like Wada) would not have been widely accepted.

History teaches us much. Among other things, we see that a conception can repeatedly arise and be fashionable only to lose acceptance again in the face of reactive Criticism, although in some cases eventually accumulating sufficient support to survive somewhat longer with each reincarnation. Even so, we know that no matter how useful a therapeutic technique is, the odds are high that it will eventually be outmoded. Meanwhile, however, callosotomy has illustrated to a notable degree the interplay among social, scientific, and clinical concerns.