Comparative Physiology of the Brain and Comparative Psychology

FIG. 39. CEREBRAL HEMISPHERES the fore-leg, in the latter dis- A, optical region; D, brachial region; tUrbattCCS Of V1S1OH. G, region of the hind-leg. (After Jj- J f COUrSC, HOt tO be expected that the distri- bution of segmental fibres on the cortex follows min- utely the arrangement of the ganglia in the spinal cord. Displacements of elements occur in the cerebral hemispheres during the process of growth. This is indicated by the formation of folds formed on the surface. It is possible that not all the segmental ganglia send fibres directly to the hemispheres, and it is possible that certain ganglia are connected with the cortex at more than one spot or region. From the

fact that the different bundles of fibres from the vari- ous segmental ganglia enter at different spots in the cortex, some authors have drawn the conclusion that there is not only an anatomical localisation of fibres, but also a psychic localisation of functions. They assume that the various psychic functions take place in different regions of the cortex. The occipital region, where the fibres from the segmental ganglia of the optic nerve enter, is considered by these authors as the seat of visual consciousness. At the spot D (where the brachial fibres enter or leave) the " con- sciousness of the fore-leg" is said to be localised. These assumptions are contradicted by the plain facts of associative memory. Simultaneous processes in different sense-organs are fused in our memory. The odour of a rose recalls its visual image. This could not be possible if the visual processes were confined to one region of the cerebral hemispheres ; they must spread to the olfactory region, and vice versa. The same can be said of other kinds of stimulation and of combinations of more than two stimuli. Moreover, we remember not only simultaneous sense-impressions, but we remember a whole series dependent upon suc- cessive stimuli of different character, if only the first constituent of such a series has been aroused. This indicates that even the after-effects of a stimulus must spread all over the cerebral hemispheres, so that they may fuse with the successive processes going on in the brain. It is thus obvious that the assumption of a localisation of psychic functions in the cortex is

opposed to the elementary facts of associative memory or consciousness. 2. Experiments on the brain indicate that while there exists to a certain extent an anatomical localisa- tion in the cortex, the assumption of a psychical localisation is contradicted by the facts. The occipi- tal region of the cerebral hemispheres is said to be the seat of visual processes, the temporal lobes the seat of auditory processes. If the occipital regions are removed, only the visual processes are said to cease, while if the temporal regions are removed, only the auditory processes are said to disappear. We know that persons who were born blind and deaf have shown a normal or even a superior intellect (Laura Bridgman). If the theory of psychic localisation were correct, we should expect that an animal from whose hemispheres the occipital and temporal regions are removed would become blind and deaf, but would remain normal in other directions. But Goltz has shown that such an animal (dog) becomes hopelessly idiotic (2, V.). The processes of association even of the other senses are no longer normal. This agrees with the idea that in processes of association the cerebral hemispheres act as a whole, and not as a mosaic of a number of independent parts.

Goltz has proved that if we remove one whole hemisphere in a dog the personality of the animal or, in other words, the sum-total of its associations remains the same. The dog recognises its friends and all the other objects it has ever known, and it reacts in such a way as to indicate that its associative memory has not suffered through the operation. But if the an- terior parts of both hemispheres be removed, the dog is no longer normal, but idiotic. It no longer reacts in the same way it did before, and it is obvious that its associative memory has suffered. The same is true if both posterior halves of the cerebral hemispheres be removed (2, V.).

If we ask at present what determines this difference, we are at a loss to give an answer. We might point out that the right and left hemispheres are practically symmetrical, while the anterior and posterior parts are not symmetrical. If the form or orientation of the elements be of importance, we might conceive of the possibility that in a brain with only one cerebral hemisphere all the processes could occur in approxi- mately the same form, while in the brain with both posterior or both anterior halves of the hemispheres gone, the processes of association could not be re- peated in the same, but in a mutilated form. Hence the idiocy which follows such operations. We might illustrate this by an analogous experience in the phy- siology of sound. Each vowel is determined by a sound of a certain pitch. If a singer sings in a pitch higher than that of the determinant, the vowel becomes indistinct. It is possible that in the brains of the above-mentioned dogs the associations are rendered impossible or difficult, because certain elemental pro- cesses are no longer possible.

servations of Goltz indicate a connection of the main regions of the cerebral hemispheres with certain regions in the medulla oblongata. A dog that has lost the anterior halves of both cerebral hemispheres has a tendency to run with its head bent down. A dog which has lost the posterior halves of both hemi- spheres shows the opposite tendency. It moves very little and its head is carried high in the air. Its an- terior legs are stiff and often stretched forward. The difference in the position and progressive motions of these two animals seems to be somewhat similar to the difference in the attitude of an Amblystoma when stimulated by constant currents. The position of the dog in which the anterior halves of the cerebral hemispheres are removed resembles that of an Am- blystoma in a descending current, while the attitude of the dog without the occipital halves of the hemi- spheres is like that of an Amblystoma in an ascending current. (See Chapter XI.) If in a dog one cerebral hemisphere be removed, while the other is intact, the dog makes circus-motions toward the injured side. There is an unmistakable analogy between these ob- servations and the older experiments of Magendie and Flourens on the sectioning of the crura cerebelli. While dogs after the loss of the anterior halves of the cerebral hemispheres often become irritable and ugly, dogs which lose the occipital halves of both hemispheres invariably become good-natured and harmless. This indicates a connection of the cerebral hemispheres with organs of the body for which with

our present knowledge of anatomical localisation we cannot yet account. 4. Those who believe in a psychic localisation in the cerebral hemispheres base their claims chiefly on the effects of small lesions. If our point of view is correct, we should expect that small lesions either make no noticeable functional disturbance at all, or cause disturbances which are no more psychic than those following the cutting of a peripheral nerve. Hitzig and Fritsch were the first to destroy the cortex of the centre of the fore-leg (D, Fig. 39) in one of the hemispheres of a dog (i). When this centre was destroyed in the left hemisphere, the right leg showed the following disturbance : " In running, the animals did not use the right fore-paw to advantage. It was turned in or out too much and did not furnish a proper support. This never happened with the other paws. Movement did not fail entirely, but in the right leg the movement of adduction was somewhat weaker. In standing, the dorsal side of the paw was often used in- stead of the sole " (i). If the paw was placed in abnor- mal positions, no attention was paid to it by the dog. Hitzig and Fritsch draw the following conclusions from these observations : " The animals evidently had only an imperfect conscioiisness of the condition of this limb ; they had lost the ability to form perfect ideas concern- ing it" In the opinion of Hitzig we have to deal with a psychical disturbance, or, as we should say, a disturbance of associative memory. This disturbance of associative memory is, however, confined to such

processes as involve the right fore-leg. In our opinion, the phenomena observed by Hitzig are the outcome of a weakening of certain groups of muscles and a diminu- tion of the sensibility in the right leg. Such disturb- ances could just as well be produced by a pressure upon certain peripheral nerve-fibres. That Hitzig's psychological interpretation of his observation is wrong has been proved by Goltz. If Hitzig's idea were correct, we ought to assume that, if the centre of the right fore-leg were removed, a dog should no longer be able to use the right paw as a hand, where such a use is based upon the activity of associative memory. Goltz not only removed the centre but the entire left hemisphere of a dog that had been taught to dig its food out of a heap of pebbles. This dog showed all the disturbances of the right leg which Hitzig described. Yet it con- tinued to dig its food (pieces of meat) out of the heap of pebbles with the right fore-paw. It preferred to use the left paw for this purpose. But when this was forbidden it used the right paw with success. This experiment proves that the conscious or psy- chical character of the motions of the fore-leg is not affected by the removal of its cortical centre. A close observation of the way the dog uses this paw shows that certain muscle-groups must have suffered by the operation, and a closer analysis of these purely muscular disturbances explains the anomalies which Hitzig had mistaken to be of a psychical character. Removal of the fore-leg centre causes a decrease in

the tension of the extensors of the leg (and perhaps also of other groups of muscles). For this reason the leg slips easily and bends in the ankle-joint so that the animal sometimes steps on the back instead of the sole of the foot. It does not notice if the leg is placed in an abnormal position. This is partly due to the diminution in resistance caused by the weakening of certain muscles and partly due to a reduction in the sensibility of the skin. Goltz has proved that it re- quires a greater pressure on the skin of this leg to cause the dog to withdraw it than on any of the other legs. This explains also why the dog does not notice if the foot whose cortical centre has been removed is placed in cold water. There are, then, changes in the tension of certain muscles and a reduction in the sensibility of the skin which suffice to explain all the disturbances observed by Hitzig, but there is no loss of muscular "consciousness" as Hitzig assumes. To a certain extent, similar effects can be produced by dividing the posterior roots of the arm-nerves. It would hardly occur to anyone to maintain for this reason that the psychic centre of the arm-movements is localised in the posterior roots. Further proof that these disturbances described by Hitzig are due to a decrease in the tension of the extensors is furnished by the fact that in man, when an arm becomes para- lysed after a local disease in the cerebral hemispheres, a contraction producing a flexed position of the arm takes place after a time. Not all the muscles of the arm are completely paralysed as a result of the disease

in the hemispheres ; but the tension of the extensors has decreased, and as a result the tension of the flexors alone determines the position of the arm. In Goltz's experiment the centre of the right fore- leg alone had been removed. It has been said that the centre of the left fore-leg situated in the other hemisphere performed the psychic functions for both legs after the operation. I made an experiment to which this objection is not possible. A dog was taught to walk on its hind-legs when it wanted to be fed. Then the hind-leg centres were removed (G, Fig. 39) in both hemispheres. In spite of this loss the dog was still able to walk on its hind-legs. When- ever I offered it food or whenever it expected to be fed it rose voluntarily on its hind-feet. The conscious actions or associations for the use of the kind-legs liad not suffered, but there was decidedly a muscular dis- turbance inasmuch as the dog was not able to stand so long on its hind-legs as it could before the opera- tion. I showed this dog at the naturalists' meeting in Berlin in 1886. The day after the demonstration I showed the brain of the animal that had been killed in the meantime. The hind-leg centres had been removed completely.

It must, however, be explicitly stated that not every limited lesion in the motor centres leads to a disturb- ance. This is not only of importance from a theoretical but also from a practical point of view. A physician need not be surprised if a post-mortem examination shows a circumscribed lesion in the cortex which had not caused any clinical symptoms. It is obvious that certain organs are more easily disturbed by a lesion in the cortex than others. An operation in the centre of the fore-leg produces disturbances more easily than an operation in the centre of the hind-leg. There are certain parts of the body in which no disturbances can be produced by the extirpation of their so-called centres in the cortex. Nobody has thus far been able to produce a paresis or paralysis of the upper eyelid in a dog or to produce loss of sensibility in the cornea by an operation in the cortex. It must, moreover, not be overlooked that all the disturbances which follow small lesions of the cortex in dogs are only transitory.

5. Not only the motor but also the sensory disturb- ances which follow an operation in the cerebral hemispheres have been interpreted as psychic dis- turbances. We know that a lesion of the surface of the occipital lobes causes visual disturbances. Munk has interpreted these disturbances which follow a small lesion of one of the visual spheres as psychic (4). There is a small region (Aj Fig. 39) in each of the occipital lobes the destruction of which, according to Munk, causes a psychic blindness in the opposite; eye. By psychic blindness Munk means the fact that the dog does not recognise what it sees, although it is by no means blind. If the cortex be removed at the region A1 in the left hemisphere, the dog shows psychic blindness in the right eye. Such a dog, for instance, is no longer afraid of a burning match or of the whip,

provided its left eye be closed. Munk assumes that the image of memory of the whip or the burning match had been deposited in the region A, and was lost with the loss of this place. It can be shown that Munk is as much mistaken concerning the psychic character of the visual disturbance following the de- struction of a small region in the occipital lobes, as Hitzig was in regard to the psychic character of the motor disturbances following the destruction of a centre of the fore-leg. In the majority of cases the removal of the region At in one hemisphere produces no visual disturbance. In the cases where a visual disturbance is produced it is only temporary. I noticed indeed that such dogs may no longer recog- nise objects in the opposite eye, but the reason for this is altogether different from that assumed by Munk.

It is known that in man the destruction of the visual sphere in one hemisphere causes the same dis- turbance as the destruction of the optic tract of the same side — namely, a hemianopia of the opposite half of the visual field. This disturbance is not psychic but purely physiological, inasmuch as it results in a loss of irritability on one side of each retina, but not in a loss in the processes of association. The same occurs in a dog whose visual sphere has been injured in one spot, with this difference, however, that the loss of irritability is not complete. Thus if the left occipital region be injured in a man, a hemianopia of the left sides of both retinae follows, and the patient

sees nothing in the right half of his visual field. If the same operation be performed in a dog, it causes not a complete hemianopia but a hemiamblyopia (5). The dog is not blind for the right half of its visual field, but has only a reduced power of vision. It be- haves like an animal that pays less attention to that half of its visual field, or whose threshold for this half is reduced. If we stand before such a dog and hold two pieces of meat in front of it, simultaneously, one piece in each hand, the dog invariably chooses the piece at its left. It almost seems as though it did not see the piece at the right. Now we know that a moving object acts as a stronger optical stimulus than a stationary object. If the two pieces of meat are again held before the dog in the manner described above, only with the difference that the piece that is in the right half of the field of vision is moved, the clog jumps at the latter (6). This proves that in the dog the threshold of stimulation for optical stimuli has been raised in the right half of the field of vision. But how could Munk mistake the hemiamblyopia for a psychic disturbance? In a dog, the divergence of the optical axes is greater than in man. Hence the right half of the visual field is controlled more by the right eye than by the left. If we produce the hemi- amblyopia or the hemianopia in a dog, the eye oppo- site the injured hemisphere is blind or injured for considerably more than one half of its retina. If the other eye of such a dog be closed, its field of vision is reduced to a very small area, and the dog does not

recognise the objects, although it is not entirely blind. What Munk mistook for psychic blindness is, in real- ity, only hemiamblyopia or hemianopia (5, 6). 6. One of the main arguments which Munk used for his assumption of a psychic character of the visual disturbances caused by the effects of a unilateral lesion of a visual sphere was the fact that these disturbances disappear in about six weeks. On the basis of this fact he constructed the following hypothesis : The visual images of memory are deposited each in a single ganglion-cell or a group of cells in the region At of the opposite hemisphere. If this region be re- moved, the dog loses all its images of memory. But new images of memory can be deposited in the sur- rounding parts A. This will be done after the loss of the region Aj and the dog becomes normal again after six weeks. If this hypothesis of Munk were correct, visual disturbances of such a dog should not disap- pear if it were kept in the dark, where it would have no chance to acquire new visual images of memory. I made that experiment. In dogs which possessed only the right eye, the region Aj in the left cerebral hemisphere was destroyed. In the majority of these dogs, the operation produced no effect. In a few, hemiamblyopia occurred. Of these several were put in an absolutely dark room for the following six weeks. As soon as they were taken out they were entirely normal. This proves that their recovery was not due to the acquisition of new visual images of memory, but to the fact that a purely physiological effect upon

the irritability of the optical apparatus caused by the operation wears off after a certain time. We then come to the conclusion that the apparent psychic blindness which follows the destruction of the region Aj in the opposite hemisphere is exclusively a hemianopia or hemiamblyopia. This disturbance is no psychic disturbance inasmuch as it can be pro- duced by an injury to a peripheral nerve, the optic tract. 7. We must seek an explanation for the temporary character of the disturbance which follows small lesions. If the lesion covers a large area, the dis- turbance is more permanent. Goltz assumes that these transitory effects are shock-effects due to the operation. He was led to this assumption through his experiments on the spinal cord. If the spinal cord be cut in a dog, no segmental reflexes occur during the first days or weeks after the operation in the part of the animal below the cut. " Pressing on the hind- feet produces no reaction. In the male dog, erection of the penis cannot be aroused reflexly. The urine collects in the relaxed bladder. The anus gaps. In brief, the whole posterior part of the body seems un- irritable. A few days later the apparently dead spi- nal cord may have recovered almost entirely. The posterior part of the animal then offers a large num- ber of reflex phenomena. No one will assume that that piece of the spinal cord which is separated from the brain in so short a time acquires entirely new powers as a reflex organ ; we must assume that these

powers were only suppressed or inhibited temporarily by the lesion of the spinal cord." The same is true in regard to the vasomotors. Division of the spinal cord reduces the tonus of the blood-vessels of the posterior legs. After a time the blood-vessels recover and become normal again. Now if the sciatic nerve in the same animal be severed, a new temporary para- lysis of the vasomotors follows. This proves that the vasomotor paralysis in the hind-legs that occurs after the division of the spinal cord is due to a shock-effect of the operation. What the nature of this shock-effect is we do not know. Perhaps v. Cyon's experiment throws light on this : v. Cyon showed, namely, that the tension of the muscles de- creases after the division of the posterior root of their segment (3).

8. We conclude from all these observations on dogs that small lesions do not cause any disturbances in the processes of associative memory, and that Hitzig and Munk are wrong in interpreting the disturbances fol- lowing the excision of a small piece of the cortex as psychic disturbances. In the majority of cases such slight lesions cause no disturbance, and where any is caused it is of such a character as could be pro- duced by the lesion of a peripheral nerve. If we wish to produce psychic disturbances by a lesion of the brain, we must destroy extensive parts of both hemi- spheres. Operations in one hemisphere alone, and even the destruction of an entire hemisphere, have no such effect.

It has been claimed that the intellect is the func- tion of special parts of the brain. Hitzig and others assumed that the frontal lobes of the cerebral hemi- spheres are the organs of attention. I have repeat- edly removed both frontal lobes in dogs (6). It was impossible to notice the slightest difference in the mental functions of the dog. There is perhaps no operation which is so harmless for a dog as the re- moval of the frontal lobes. Flechsig thinks that it is not only the frontal lobe but the cortex of certain other regions which is responsible for mental activity, inasmuch as it is the seat of " centres of association." I have removed the cortex of Flechsig's " centres of association " in dogs without having noticed anything that justifies Flechsig's hypothesis. The assumption of " centres of association " is just as erroneous as the assumption of a centre of coordination in the heart. Association is, like coordination, a dynamical effect determined by the conductivity of the protoplasm. Associative processes occur everywhere in the hemi- spheres (and possibly in other parts of the brain), just as coordination occurs wherever the connection be- tween two protoplasmic pieces is sufficient. It is just as anthropomorphic to invent special centres of associ- ation as it is to invent special centres of coordination.

1. HITZIG, E. Untersuchungen iiber das Gehirn, Berlin, 1874 ; and Reicherfs und Du Bois-Reymond's Archiv, 1870. 4. MUNK, H. Ueber die Functionen der Grosshirnrinde. Ber- lin, 1881. 5. LOEB, J. Die Sehstorungen nach Verletzungen der Gross- hirnrinde. Pfliigers Archiv, Bd. xxxiv., 1884. 6. LOEB, J. Beitrdge zur Physiologic des Grosshirns. P finger's Archiv, Bd. xxxix., 1886. i. We have mentioned the hypothesis that each image of memory is localised in a special ganglion-cell or a group of ganglion-cells. As soon as a new image of memory arrives, it is, according to this hypothesis, deposited in one of the empty cells. Who deposits it and who finds out which cell is empty and which oc- cupied is a question the originators of such hypotheses do not ask. This conception treats the image of memory as if it were something substantial, i. e., something characterised by mass.1 Munk has as- serted the possibility of proving that in a dog the sin- gle visual images of memory are localised in isolated cells, or groups of cells, at the part Ax (Fig. 39). He gives as proof two experiments, " in which extirpation of the part Aj caused the loss of all but one of the visual images of memory. One single visual image

1 This peculiar hybrid between metaphysics and anatomy owes its origin largely to Gall. Gall was an industrious worker in the anatomy of the brain and at the same time a huge fraud. The anatomy of the brain was not suffi- ciently sensational for him, so he enlivened things somewhat by grafting upon his anatomy the worst metaphysics he could possibly get hold of. The various nooks and corners of the brain became the seat of soul-powers of his invention. This artificial connection between metaphysics and brain-anatomy or histology has since become traditional.

of memory in each case was found to be preserved and unimpaired : in one case the image of the pail, out of which the dog was accustomed to drink, remained ; in the other, that of the motion of the hand, which be- fore the operation had been the signal for the dog to give its paw." It was this statement of Munk that led me as a student to make experiments on the brain. I hoped that a road to an exact psychology had been opened. I began my experiments as a con- firmed supporter of Munk. The more experiments I made the more it became apparent that many of Munk's statements were incorrect, especially his mea- gre statements concerning the supposed localisation of single images of memory. It is my opinion that these histological or corpuscular hypotheses of the images of memory must be supplanted by dynamical conceptions. The dynamics of the process of associ- ation is the true problem of brain-physiology. Even if the hypotheses of psychic localisation were not contra- dicted by all the facts, the pointing out of the centres would not be a solution of the dynamical problem. By merely showing a student the location of a power- plant, we do not explain to him the dynamics of electric motors.

I have mentioned above the possibility that pro- cesses of association will become abnormal if certain elemental constituents are mutilated or impossible. I selected as an example our ability to recognise a vowel. If the vowel is sung at a pitch which excludes its specific formative sound, it becomes indistinct. A study of patients afflicted with amnesia seems to support this analogy. It is not possible to use all the reports of such cases. I think that the majority of practitioners have neither the training nor the time to analyse them. I will confine myself to two cases from the Clinic of Professor Rieger in Wiirzburg, one of which was analysed by himself (i), and the other by his assistant, Dr. Wolff (2). In the first case the patient had suffered a concussion of the brain in a railroad accident. Among a number of other disturbances, his memory showed peculiar gaps. The patient was able to recognise only the numbers i, 2, and 3. The corpuscular theory of the images of memory would assume that all numbers which the pa- tient had originally possessed had been located each in a special cell, and that these cells had all perished with the exception of the cells which contained the first three numbers. This at once seems strange ,and becomes still stranger when taken in connection with the following observation. In every case it took the patient some time to find the word one when the fig- ure i was held before him. The reaction-time for naming a 2 was considerably longer, and for naming a 3 was still longer. He was able to reckon with these three numbers, but when a 3 occurred he required more time than when a i or a 2 occurred. The determin- ation of the reaction-time furnishes the explanation of the fact that all numbers beyond 3 were wanting. All of Rieger's experiments on this patient showed that if he did not succeed in finding the name of an object

within a certain time (about eighteen seconds) it was impossible for him to do so at all. Now for finding the word three when he was shown the figure 3 he required almost eighteen seconds, and in fact he even failed occasionally to find it. The first three numbers are the ones that a child first learns, and are also those used most frequently during life. We know that the words we use least are the ones most liable to vanish from our memory (for instance, the vocabu- lary of a foreign language). It is possible that in the brain of the patient the processes were partly mu- tilated or rendered more difficult. The numbers used most frequently could cross the threshold ; those used less frequently could not. This conception is further confirmed by the fact that by touching the edges the patient was able to distinguish a ten- from a fifty-pfennig piece, although the numbers ten and fifty were otherwise gone, and as stamped on the coins were only hieroglyphics to him. The money- conception of the ten- and fifty-pfennig piece had formed more associations and clung more tenaciously to the memory of this man, who had to struggle for his existence, than the abstract conceptions ten and fifty, which had existed in his head only as a scholas- tic luxury. Hence any adequate idea of the nature of the disease of this man must be a dynamic one. In the injured brain of this patient certain processes were able to take place as before, except that they were less intense or incomplete. Those innervations forming constituents of relatively many or important

associations were still possible, or occurred in a more normal form, while other innervations became impos- sible or were mutilated. In this case it would be just as erroneous to assume that the single conceptions or letters are all localised in single cells, and that the corresponding cells in the patient had perished, as it would be erroneous to conclude in a case of interfer- ence of sounds that the source of vibration was removed.1 2. The second case mentioned is still clearer (2). The disturbance of associative memory was also caused by an accident. When the patient was asked the colour of the leaves of a tree, he was unable to answer the question unless he was allowed to go to the window and look at a tree. In this case he an- swered correctly. As long as he could not see a tree it was impossible for him to tell the colour of a leaf. Pieces of green, red, and blue paper were put before him, and he was asked which the leaves looked like, but he was unable to tell. If asked whether the trees were blue, he answered that this was possible. Only when looking at a tree was he able to remember that the leaves were green. When asked how many legs a horse has, he went to the window and waited until a horse passed by. This enabled him to find the word four. Only in winter was he able to tell the colour of snow. In summer he admitted the possibility that

1 Conditions similar to those that existed in this patient can be artificially produced by the dynamometer-experiments which will be described in the next chapter. snow is black. He was once asked the colour of the blood. He opened a little pustule on his hand, and as soon as a drop of blood came out he gave the answer, red. It is obvious from these facts that the patient un- derstood every question and was sufficiently intelli- gent to secure those impressions which allowed him to answer the question. He could tell the colour of sugar if allowed to look at it, but this did not help him to tell whether or not sugar tastes bitter. In order to do so he had to put the sugar into his mouth. When a smooth piece of glass was shown to him he could not tell that it was smooth until he had touched it.

Two things are evident — first, the patient was not able to remember any perceptible quality of an ob- ject unless the object was under his immediate per- ception ; and second, he remembered the various qualities only if the specific senses for these qualities were affected. In a normal being the word siigar or the sight of sugar suffices to produce the association of its sweet taste. In this patient only contact with the tongue suggested the word sweet, although he was intelligent enough to know how to arouse the correct association.

The names of a great many objects may be sug- gested to a normal person through any of several of the senses. For example, we find the word violin if we see the object as well as if we hear it played with- out seeing it. The patient in this case was a violin player, hut it was necessary for him to see the instru- ment before he could name it. When a key was put into his pocket and he was allowed to touch it he could not say what it was. He could, however, find the word if he could see the key in the door. When his hand was put to his ear, he could not tell what he touched unless he looked at the doctor's ear. When the doctor covered his own ear, the patient was un- able to find the word. It is obvious that in his case the visual perception was, on the whole, more effect- ive than any other sensation. A sense-perception was necessary to call forth the association of concrete objects, and of the many possible sense-perceptions, which in normal cases mi<rht have brought about the

word, the strongest alone in him sufficed. The word umbrella was only suggested when the umbrella was opened. From this we might imagine that a change in the machinery of association had taken place, which allowed only the processes having a maximal intens- ity or amplitude to arouse an associative process, the others remaining without any effect. The process was the same in regard to abstract as- sociations. The patient complained that his annuity was too small. He remonstrated against the doctor's insinuation that he had murdered his wife or that he was a scamp. But whether a beggar is a wealthy or a poor man, or whether God lives in hell or in heaven, were problems which he was not able to de- cide, although he was a believer.