Brain (1984), 107, 155-167 COLOURED PATTERNS IN HEMIANOPIC FIELDS by HANS W. KOLMEL (From the Department of Neurology, Klinikum Charlottenburg, Freie Universitat Berlin, Spandauer Damm 130, 1000 Berlin 19) SUMMARY In a group of 96 patients with homonymous hemianopia or quadrantanopia from various causes, usually vascular lesions, 14 experienced coloured patterns in the hemianopic field. A questionnaire was used in an attempt to obtain as much information as possible on these phosphenes. They appeared in four major colours—red, green, blue and yellow—usually infieldsthat joined to form horizontally and vertically orientated geometric patterns. They demonstrated translocation related to ocular movement but otherwise were not influenced in any way by ocular movement or by opening and closing the eyes. The patients experiencing these phosphenes constituted a homogeneous group in relation to neurological findings, cause of hemianopia, CTfindingsand prognosis. AH had sustained cerebral infarction, and CT studies revealed small circumscribed lesions, usually subcortically in the striate area of the interhemisphericfissureand the adjacent white matter in the occipital lobes. The visual defects resolved completely or to a considerable extent in many cases, which suggests that the phosphenes are associated with hemianopia due to a functional disorder rather than to tissue loss. Coloured patterns are therefore a prognostically favourable symptom. The geometric structure of the phosphenes and the fact that the same four colours consistently reappear must reflect the neuronal organization of the visual cortex. It can be postulated that this functional unit is organized on the basis of pyramids and tetrahedrons. The similarity of the geometric structures to the fortification lines encountered in migraine raises the possibility that the patterns are generated in the striate cortex. Absence of alteration in size with increasing distance from the fovea suggests that these phosphenes originate in a prestriate area. The colours reported indicate stimulation of intact colour-coding neurons in the prestriate area. INTRODUCTION A number of patients with homonymous hemianopia report phosphenes in the blind visualfield,particularly in the early phase of the disorder. There are accounts of such phosphenes in the literature (Eskuchen, 1911; Horrax and Putnam, 1932; Teuber et ai, 1960; Gloning et al., 1967; Lance, 1976), but the variety of descriptions given makes a clinical or pathophysiological interpretation difficult. Systematic clinical studies of the visual phenomena in hemianopicfieldshave not so far been reported. This is not surprising, since it is difficult to impose an organizing principle on the wide variety of descriptions of coloured and colourless phosphenes. The following is an attempt at characterizing a particular type of phosphene which has been described with a high degree of uniformity by a large number of 156 HANS W. KOLMEL patients. These phosphenes may be termed coloured patterns. The phosphenes will be described in detail and their relationship to homonymous hemianopia discussed. In addition an attempt will be made to determine whether focal cerebral lesions are responsible for the appearance of these phosphenes. Finally the extent to which coloured patterns permit conclusions as to the nature and organization of the neurons generating these phenomena will be discussed. PATIENTS A total of 96 patients with homonymous hemianopia or homonymous quadrantanopia were examined over a period of five years. The distribution of visual field defects was as follows: right homonymous hemianopia, 36 patients; left homonymous hemianopia, 36; right upper field, 4; right lowerfield,2; left upperfield,9; left lowerfield,9. Most patients came to examination as a result of an acute visual disturbance. A minority (25 patients) presented with visualfielddefects dating from one to thirty-six years previously and were examined in the course of routine follow-up studies or were included on the basis of an incidental finding at neurological examination performed for other reasons. The onset of the visual disorder could not be determined in two patients. After neurological examination, ophthalmological studies were performed with special attention to the visual fields, visual acuity and foveal colour recognition. The visualfieldswere charted with a Goldmann perimeter and foveal colour recognition tested with Ishihara charts. Table 1 shows the sex distribution and the aetiology of the visual disorder for the whole group. Cerebrovascular disorders were the most common cause. Cranial computed tomograms (CT scans) were made in almost all patients in order to determine the nature and location of the cerebral lesion, with electroencephalographic studies in some cases. Angiography was performed when indicated. TABLE 1. AGE AND SEX DISTRIBUTION AND AETIOLOGY OF THE VISUAL DISORDER Sex M 61 F 35 Age at first appearance 18 to 83 years Cause 68 Ischaemic stroke Space-occupying lesion 25 3 Other METHODS Method of Evaluating the Phosphenes In a preliminary examination, information was obtained on phosphenes from patients with homonymous hemianopia, after the patient's attention had been called to the possibility of such visual experiences. These interviews revealed that detailed information on phosphenes was only obtainable when the interviewer repeatedly asked about such events. A standardized interview would have been too rigid to produce results. Therefore a questionnaire based on relevant neurological and neurophysiological knowledge was developed in order to assist the interviewer in extracting as much information as possible on these visual phenomena. This questionnaire included 14 points. (1) When did the visual field defect or disorder of vision occur and when were phosphenes first observed? (2) During what period of time were visual phonomena observed? How long did the individual sensations last? (3) How often did the phosphenes occur and did they occur regularly? (4) In which part C O L O U R E D P A T T E R N S IN H E M I A N O P I C F I E L D S 157 of the visual field were the phosphenes observed? (5) Did the phosphenes fill the entire visual field or were they limited to certain locations? Did they change their position from episode to episode or were they always found at the same location? (6) What shape did they have? (7) When more than one phosphene was observed at a time, were all similar in size? If they were of different sizes, where were the larger and smaller ones usually located? (8) Were they coloured or not? What colours? (9) Were the visual sensations similar? (10) Were single points of light or groups of points observed? When groups appeared did they seem to have a particular organizing principle? (11) Was it possible to influence the phosphenes in any way? (12) How did opening or closing the eyes influence the visual phenomena? (13) What happened to the phosphenes when the eyes were moved or when a near or distant point was fixed? (14) When were the phosphenes last seen? Most of the patients were examined with a tangent screen at a distance of 1 m in order to localize the phosphenes in the visual field and to estimate their size. The patients fixed the centre of the screen and pointed out the position of the phosphenes with the hand and described the direction of movement in cases where this occurred. This method of demonstration proved useful in determining the position of the visual phenomena. The studies using the tangent screen also allowed a much more critical evaluation of the visual sensations. The patients repeatedly demonstrated a tendency to look in the direction in which the phosphenes presumably moved. When the midpoint of the screen was fixated the patients' statements on motion and direction of motion were much less certain. Some of the patients then expressed the opinion that the phosphenes had not moved at all. The subjects were then instructed to pay special attention to motion when phosphenes subsequently occurred. The patients were also requested to note the influence of eye movements on visual effects. However, the additional information obtained revealed nothing of major importance about the phosphenes. The patients were initially asked to provide a spontaneous description of the colours observed during their visual sensations. They were subsequently requested to select the colours among the 16 different tones in Hering's (1920) colour wheel. The patients tried to estimate the size of the phosphenes on the tangent screen. Finally they were asked to make drawings of the phosphenes that they had observed. The results varied widely in quality, and some patients were incapable of drawing their visual sensations. CASE REPORTS Case 4 The patient, a man aged 72 years, was known to have had elevated blood pressure levels on occasions during the previous few years. While shaving one morning he experienced the sudden onset of dizziness and a temporary loss of vision lasting approximately one minute. After improvement in vision, spatial disorientation lasting approximately two hours was noted. The patient was free of symptoms on admission to hospital. The following night he was awakened by the sudden appearance of bright coloured spots on the left side. The figures were three-dimensional rhomboids with yellow, green, blue and red facets and appeared suspended in space. The rhomboids were joined in vertical chains, sometimes in multiple figures, and seemed to create a pattern. They always appeared in the upper portion of the visual field and descended slowly to the midpoint or below before moving to the left and disappearing. A few seconds later they reappeared at the same point and repeated their course. They always obliterated the visual field in their path. The repeated appearance of these blinking and flashing figures with their stereotyped motion disturbed the patient to such an extent that he notified the physician on call. He was seen by an ophthalmologist the next morning. The patterns were smaller when the patient fixed a near point and larger when he fixed a distant point. The patterns changed position with eye movement, but opening and closing the eyes had no effect. The rhomboids appeared to measure approximately 0.5 deg on the tangent screen, and pure colours—red, yellow, green and blue—were selected from the Hering colour wheel. The patient was able to make very exact drawings of his visual experiences. 158 H A N S W. K O L M E L FIG. 1. Coloured pattern in the as yet unaffected left visual field of Case 4. Drawing by the patient. The sides of each pyramid were coloured red, yellow, green or blue. The colouredfiguressubsequently appeared less often and disappeared completely on the third day of his hospital stay. At the same time an almost complete left homonymous hemianopia was found. There were no other abnormalities on neurological examination. CT scans revealed slight cortical atrophy and a hypodense zone in the right occipital lobe. An EEG recorded two days after disappearance of the visual phenomena revealed slight episodic activity but no focal changes. The hemianopia was unchanged. The descriptions of the phosphenes were identical one year later. Case 9 The patient, a man aged 69 years, has suffered from hypertension for many years and diabetes mellitus, treated with oral medication, for eight years. A left hemiparesis from a cerebral infarct occurred at the age of 67 years. Restitution of motor function was almost complete. Intermittent claudication developed one year later, and aortofemoral bypass surgery was performed. At the age of 69 years another vascular lesion occurred resulting in a complete left homonymous hemianopia with sparing of the macular region. The EEG demonstrated focal activity in the right temporal area, and a CT scan showed a hypodense zone in the right occipital lobe. The patient described the initial visual disorder as follows. 'I woke up in the morning feeling well and turned to look at my wife who was lying to my left in bed. AtfirstI could not see her at all. I had to turn my head far to the left to see her in bed. It was immediately clear to me that something was wrong with my vision on the left side. However, I was less disturbed by the loss of vision than by the appearance of flashes of coloured light on the left. These appeared as bright stars with all the colours of the rainbow, distributed over the otherwise dark left side; 4 to 8figuresflashedsimultaneously for a period of several seconds and disappeared only to reappear immediately in the same pattern at another location. I sometimes had the impression that an entire net of colours appeared. The visual effect was so unpleasant that I decided to see an eye doctor. COLOURED PATTERNS IN HEMIANOPIC FIELDS 159 FIG. 2. Coloured pattern in the hemianopic right field of Case 12. The facets of each element were red, yellow, green or blue in colour. The next day a coloured fog appeared a number of times for a period of several minutes on the right side. At that point I was admitted to hospital. Four days after their initial appearance the visual phenomena disappeared suddenly during an intravenous infusion.' The homonymous hemianopia persists unchanged. Case 12 The patient, a 34-year-old nurse on an ophthalmology service, had never been seriously ill until she awoke one morning at 7 a.m. with a left-sided headache. At 10 a.m. she noticed that the right visual field was limited. Severely confused, she lost her sense of orientation in her own apartment. She decided to seek professional help two days later, driving her own car and following a number of detours to arrive at the ophthalmologist's office. A right homonymous hemianopia with macular sparing was diagnosed. Neurological examination failed to demonstrate other abnormalities. CT scans with contrast enhancement revealed a slightly hyperdense zone in the left occipital region. The history, the physical findings and the fact that the neurological deficits improved continuously over a period of fourteen days with full restitution of visual capacity suggested a diagnosis of cerebral infarction. Approximately two to three hours after the patient noticed her visualfielddefect, bright silver stars suddenly appeared in the dark field, especially in the upper segment. In addition, the patient saw a group of bright coloured spheres, ellipses, squares and hexagons which appeared in the upper left of the dark field and traversed the field only to disappear suddenly as though they had passed behind a black wall. The patient observed this phenomenon 8 to 10 times per day, with individual sensations lasting for a fraction of a second repeated during a period of one to two hours. Thefigureshad four intense colours. Without hesitation she selected the four pure colours from the Hering wheel. She 160 HANS W. K O L M E L estimated the size of the individual ellipses, squares and hexagons between 2 and 5 deg on the tangent screen. She did not report an apparent change in size with increasing distance from the screen. The phosphenes moved with eye movements. They were unchanged by opening and closing the eyes. After three days the coloured patterns disappeared completely. RESULTS Of the 96 patients, 40 experienced phosphenes in the hemianopic field. They perceived several different visual phenomena during the course of their illness. Coloured phosphenes were reported by 26, 17 described phosphenes with heterogeneous colours and shapes and 14 reported visual phenomena which can be classified as coloured patterns. All 14 patients described several different colours, usually four, less often two, three orfive.They selected intense yellow, red, green and blue from Hering's colour wheel; 2 patients also selected orange and one identified violet. The colours were so intense that the patients initially closed their eyes in an attempt to shut out the light. One patient described his phosphenes as though a light as powerful as the sun had projected the colour, while another patient spoke of 'shocking' colours and a third likened the visual effect to fire, which expresses the painfulness of the visual experience. The patients were usually able to describe the shape as well as the colours of their phosphenes. Comparisons were repeatedly made with a kaleidoscope held against the sun, a snakeskin, a tortoise shell, a harlequin's suit, a brightly coloured net, a cloud of burning phosphorus, or the tread of a tyre. The colours did not appear sequentially but usually simultaneously and in a constant pattern. Two patients, however, described a bright point which emitted rays of yellow, red, blue and green TABLE 2. SYNOPSIS OF CLINICAL DATA FROM 14 PATIENTS WITH COLOURED PATTERNS H Palkxt Cue 1 Cue2 Cue 3 Cue 4 Cue5 Cue 6 Cue 7 Cue! Cue 9 C u e 10 C u e II C u e 12 C u e 13 C u e 14 58 40 71 72 51 57 77 35 69 59 63 34 62 66 si 68 40 71 72 51 57 77 54 69 61 63 34 62 66 § ^ li F M F M M M M F M M M F M F L R L L L L R R L L R R R L " L 0.5 1.2 1.0 1.0 1.2 1.0 1.0 1.0 07 10 1.0 1.2 1.0 1.0 R 0.6 12 10 09 1.2 09 10 1.0 0.7 1.0 1.0 1.2 1.0 1.0 O — occipital; P •• parietal, I — cerebral infarction 3 H si Js + + + + + + + + + + - O O O O O O + + + + + + + + + + + + + + I I I I I 1 OI O I OI O I O I O I O I O/P 1 COLOURED PATTERNS IN HEMIANOPIC FIELDS 161 light. In these cases, yellow was observed initially, followed by the other colours, which were always quite distinct from one another. Together they described a figure that was most often angular, usually in the form of a star, a honeycomb, a triangle, pyramid or tetrahedron, and rarely a round form such as a circle or an ellipse. Several patients described a pattern floating in space. Most of the phenomena contained a basic figure which almost always combined to form patterns which led to the use of the terms honeycomb, harlequin's suit, snakeskin, tortoiseshell, kaleidoscope or net. Horizontal and vertical lines predominated. It was apparently difficult for the patients to estimate the size of the phosphenes. When a tangent screen was used at a distance of 1 m, estimates of the size of basic elements ranged from 0.5 to 8 deg. The patients noted no apparent relationship between the size of the phosphenes and distance from thefixationpoint, except Case 4. In general the phosphenes did not vary in size from episode to episode in the same subject. However, basic elements were sometimes eradicated from the pattern before the entire figure disappeared. The phosphenes neverfilledan entire half-field. The coloured patterns appeared suddenly and usually did not move. Initial motion was slow and lasted only one to two seconds when it occurred. Three patients reported that the phosphenes appeared high in the blind field and descended to the midline before they disappeared in the lateral or lower part of the field. It may be assumed that this represents genuine motion, since movement of the phosphenes in the visualfieldwas independent of eye movement. Opening or closing the eyes never affected the phosphenes. The 14 patients who perceived coloured patterns suffered an acute homonymous hemianopia. Quadrantanopia was not found at the onset of the visual disorder. The phosphenes occurred shortly after development of the hemianopia. They were observed simultaneously with the visual field defect in 6 patients and between one and three hours after the initial deficit in 3 other patients. Five subjects reported phosphenes before development of the hemianopia. Case 4 described quite impressively the manner in which the phosphene—similar to the scintillating scotoma experienced in ophthalmic migraine—seemed to impose itself between the viewer and objects in such a way that he was not able to see behind the phosphene. Visual capacity was restored when the phosphene disappeared. This patient and two others (Cases 1 and 14) reported that the phosphenes disappeared with the development of the visual field defect. In most cases the phosphenes disappeared within ten days of the onset of hemianopia, though they lasted fifteen days in one patient (Case 10). Fig. 3 summarizes the period of time over which coloured patterns were observed, as well as the temporal relationship of these phosphenes to the onset of hemianopia. Ten of 14 patients experienced improvement in the visual defect, usually within four weeks. Of the 3 patients in whom appearance of phosphenes preceded the onset of the visual disorder and disappeared with the development of hemianopia (Cases 1, 4 and 14), 2 reported no improvement in the hemianopia while one (Case 4) showed slight improvement in the upper quadrant. Vision returned to normal in HANS W. KOLMEL 162 3 patients (Cases 2, 11 and 12), and 4 others (Cases 3, 6, 10 and 13) showed significant improvement (fig. 4). Visual acuity was not affected by the disorder, and there were no derangements of colour description, colour perception or foveal colour discrimination. No patient suffered a neuropsychological disorder such as aphasia, alexia, agraphia, prosopagnosia or spatial disorientation. A vascular disorder in the occipital lobe was found in all patients. CT studies were made in 11 patients, and infarction in the occipital lobe was demonstrated in 9 of these. The CT appearances failed to reveal changes which might explain the clinical findings in 2 patients (Cases 2 and 12), both of whom demonstrated complete recovery of vision. The hypodense zones were small in the majority of patients, and the prestriate cortex was not affected visibly in a single case. The lesions were more often at a subcortical level affecting the optic radiation. With the exception of one patient (Case 14) who also had formed hallucinations, the cortical lesions were limited and only involved a small portion of the primary visual cortex. Case No. 3 2 — Days J1O8642 246S10 \— Hours | Hours —H Ho 1 2 3 4 5 6 7 Days - Onset of hemianopia FIG. 3. Period over which coloured patterns were observed and relation to the onset of homonymous hemianopia. DISCUSSION The characteristics of the coloured patterns may be summarized as follows. They appear in four major colours—red, green, blue and yellow—usually in fields that join to form horizontally and vertically orientated geometric patterns. They appear suddenly, usually only for a fraction of a second, and remain in one position, although they may move slowly and stereotypically within the hemianopic field in a minority of cases. Case 8 Case 2 O O Case 9 rO C 7* m O 2 days Case 3 Case 10 17 days Case 4 54 days Case II 3 days H H tn 7> 2 on 12 days X Case 5 m 2 Case 12 97 days Case 6 12 days z o n Case 13 43 days 30 days m r a Case 14 Case 7 to 28 days lday FIG 4 Visual fields in 14 patients experiencing coloured patterns. Left, visual fields at onset of the disorder. Right, ultimate shape of the visual field defect and time after which no further change occurred. (Detailed analysis of macular vision was not performed.) 164 HANS W. KOLMEL The reports by other authors on phosphenes in hemianopic fields are not detailed enough to permit adequate comparison. It is likely that the phosphenes reported by Higier (1894) and Eskuchen (1911), by Gloning et al., (1962) in a series of patients with brain infarction, and by Mooney et al., (1965) in a study on the recovery phase of homonymous hemianopia, are identical with our series. Case reports by Wilbrand and Saenger (1917) documented comparable phosphenes. For example, one 40-year-old patient with cerebral infarction 'complained of strong visual effects which appeared asflickeringflames,zig-zags, serpentine lines and lightning flashes which caused him considerable discomfort. The visual effects were strongest immediately after the cerebrovascular accident and lasted approximately 10 to 12 days.' Similar phenomena are apparent in the case reported by Kinsbourne and Warrington (1963). A patient with homonymous quadrantanopia secondary to brain infarction in the occipital lobe experienced bright multicoloured flashes of light which were at times without apparent order but which at others resembled fortification figures. Krause (1924) described patients who experienced coloured patterns in the hemianopic field for a short period following resection of occipital lobe tumours. None of the phosphenes provoked by electrical stimulation of the occipital lobes resemble the coloured patterns described above (Foerster and Penfield, 1930; Penfield and Perot, 1963; Brindley and Lewin, 1968; Dobelle et al., 1979). The phosphenes usually appear at the onset of hemianopia, but may precede the event by hours or, in some cases, days. When phosphenes occur a functional disorder of the entire visual cortex or a portion thereof, may be assumed, rather than an irreversible tissue defect. It implies that restitution of visual capacity is possible as long as the phosphenes are observed. If the coloured patterns disappear before vision improves—recognizable on the basis of grey tones in the blind field or perception of crude movement—then major visual improvement is unlikely; the visual disorder is secondary to a persistent tissue defect. Twelve of the 14 patients perceived coloured patterns in the hemianopic field; 10 of them showed an improvement of the visual field defect. The control group consists of those patients who also developed a homonymous hemianopia from a stroke but who did not report phosphenes in the hemianopic field. This comprises 38 patients, of whom only 16 showed improvement in the visual field defect. These results underline the favourable prognosis of coloured patterns. Coloured patterns seem to occur mainly in the course of a cerebrovascular event. Similar phenomena have been reported in cases of tumour, abscess or craniocerebral injury in the occipital region, but these reports usually involve visual manifestations in epileptic seizures rather than isolated phenomena in the hemianopic field. Case reports by Westphal (1881) and by Henschen (1923) emphasize this point. The significance of coloured patterns in localizing lesions is based on the results of clinical examination as well as on CT findings. Homonymous hemianopia without other neurological or neurophysiological deficits after the first four hours suggest a lesion located in the posterior portion of the optic radiation or in the area striata. COLOURED PATTERNS IN HEMIANOPIC FIELDS 165 CT studies revealed that most lesions were in the terminal part of the optic radiation, especially within the region of the area striata. The occipital cortex deep in the interhemispheric fissure and occipital convexity were usually not noticeably involved in CT studies (fig. 5). The extent to which the particular characteristics of the coloured patterns, their strict geometric figures, and conspicuous colours allows conclusions as to the cortical area in which they originate, above and beyond the information provided by clinical examination and computed tomography, will now be considered. The four colours most often described are the 'primary colours' which form the basis of colour codification in Hering's (1920) theory. It may be assumed that the appearance of these colours in phosphenes without an adequate external stimulus indicates stimulation and/or inhibition of the same neurons at a corresponding presentation in the visual hemifield. It could be suggested that neuronal stimulation occurs generating the colours in the phosphenes, especially in area V4 where Zeki (1973) found a conspicuously large number of colour-coded cells in his experiments on monkeys. Unfortunately it is as yet unclear whether these topographic and neurophysiological findings, which have been subject to recent criticisms (Schein et al., 1982), are applicable to the human brain. For example, area 17 (VI) in man is located for the most part in the occipital portion of the interhemispheric fissure, while the same area extends well over the cortical convexity in the monkey. If it is assumed that the colour-coded neurons in area V4 described by Zeki (1973) are in fact responsible for colour perception and that these results are valid for the human brain, then it follows that a lesion in these neurons would cause a corresponding deficit. However, the lesions described by Damasio et al. (1980) in their two patients with acute hemiachromatopsia or complete achromatopsia are located in the lingual and occipitotemporal gyri, rather far from the area corresponding to V4 in the human brain. In an attempt to explain this discrepancy Damasio and coworkers (1980) postulated that colour-coded neurons are concentrated in a wide belt within the prestriate area. As far as phosphenes are ri \ I \ t I i / ; i i;: V FIG. 5. Superposition of computed tomograras from 9 patients with coloured patterns in the hemianopicfield.All zones of lesions projected in the left occipital lobe. (The illustrations contain the slices which represent the largest portion of the occipital lobes (McAuley and Ross Russell, 1979). Slices through the pineal region, as well as 10 and 20 mm higher.) 166 HANS W. KOLMEL concerned, these results suggest that the colour phenomena result from stimulation of colour-specific neurons in a large area of the visual association cortex. It is as yet unclear why the colours appear in such strictly geometric patterns, since experimental studies have revealed that colour-coded neurons are nonorientated. Hubel and Wiesel (1968, 1972) did find orientated receptive neurons in the striate and prestriate area. It might be concluded that the initial stimulus generating coloured patterns originates in geometrically organized orientated receptive neurons in the striate or prestriate cortex with subsequent extension to colour-coded neurons in the prestriate cortex in accordance with strict principles governing perception of the four primary colours. The fact that phosphenes may appear before the onset of hemianopia and that they represent prognostically a favourable symptom support the assumption that a large number of intact and functionally competent neurons must remain within the striate and prestriate cortex. Another morphological characteristic of the coloured patterns deserves special emphasis. The vertical boundaries of the pyramids or tetrahedrons have striking similarity to the fortification lines observed in migraine attacks. The latter are interpreted as resulting from patterns of stimulation and inhibition in the field of macular representation extending over the cortex in area 17. Richards (1971) connected hypothetically the tips of the fortification lines and concluded that the resulting hexagonal patterns corresponded to the functional architecture of the striate cortex. The coloured patterns described by our patients suggest a functional neuronal architecture based on pyramids or tetrahedrons. The fact that visual phenomena in migraine increase in size as they wander from the fovea to the periphery of the visualfieldsis explained by differences in size of the various receptivefields(Cowey and Rolls, 1974; Dobelle et al., 1979). If it is assumed that the stimulus for coloured patterns as fortification lines originates in area 17 then it is surprising that the patients described the patterns as strictly regular without any change in size within the individual figures (pyramids or tetrahedrons). It would be expected that the individual figures are small within the cortical area of foveal representation but increase in size with increasing distance from the fovea. The patients did not describe such changes in size. However, since the coloured patterns appeared for a fraction of a second, observations of this nature would require special skill. If the pyramids or tetrahedrons fail to demonstrate an increase in size from the centre to the periphery of the visualfield,as the patients reported, then the visual phenomena might be generated by groups of neurons within an area which is not subject to the special properties governing area 17. REFERENCES BRINDLEY G S, LEWIN W S (1968) The sensations produced by electrical stimulation of the visual cortex. Journal of Physiology, London, 196, 479-493. COWEY A, ROLLS E T (1974) Human cortical magnification factor and its relation to visual acuity. Experimental Brain Research 21, 447-454. C O L O U R E D P A T T E R N S IN H E M I A N O P I C F I E L D S 167 DAMASIO A, YAMADA T, D A M A S I O H , C O R B E T T J , M C K E E J (1980) Central achromatopsia: behavioral, anatomic, and physiologic aspects. Neurology, New York, 30, 1064-1071. DOBELLE W H, TURKEL J, HENDERSON D C, EVANS J R (1979) Mapping the representation of the visual field by electrical stimulation of human visual cortex. American Journal of Ophthalmology, 88, 727-735. ESKUCHEN E (1911) Ober halbseitige Gesichtsfeldhalluzinationen und halbseitige Sehstorungen. Inauguraldissertation, Heidelberg. FOERSTER O, PENFIELD W (1930) Der Narbenzug am und im Gehirn bei traumatischer Epilepsie in seiner Bedeutung fur das Zustandekommen der Anfalle und fur die therapeutische Bekampfung derselben. Zeitschrift fur die Gesamte Neurologic und Psychiatric, 125, 474-572. GLONING I, GLONING K, H O F F H (1967) Ober optische Halluzinationen. Eine Studie an H a n d von 241 Patienten mit autoptisch oder chirurgisch verifizierten Lasionen des Okzipitallappens und seiner Grenzgebiete. Wiener Zeitschrift fur Nervenheilkunde undderen Grenzgebiete, 25, 1-19. GLONING I, GLONING K, TSCHABITSCHER H (1962) Die occipitale Blindheit auf vaskularer Basis. Albrecht von Graefes Archiv fur Ophthalmologie, 165, 138-177. HENSCHEN S E (1923) Vierzigjahriger Kampf um das Sehzentrum und seine Bedeutung fur die Hirnforschung. Zeitschrift fur die Gesamte Neurologic und Psychiatric, 87, 505-535. HERING E (1920) Grundzuge der Lehre vom Lichtsinn. Berlin: Springer. HIGIER H (1894) Ober unilateral Halluzinationen. Wien Klinik, 20, 139-170. HORRAX G, PUTNAM T J (1932) Distortions of the visual fields in cases of brain tumour. The field defects and hallucinations produced by tumours of the occipital lobe. Brain, 55, 499-523. HUBEL D H, WIESEL T N (1968) Receptive fields and functional architecture of monkey striate cortex. Journal of Physiology, London, 195, 215-243. HUBEL D H, WIESEL T N (1972) Laminar and columnar distribution of geniculo-cortical fibers in the macaque monkey. Journal of Comparative Neurology, 146, 421-450. KTNSBOURNE M, WARRINGTON E K. (1963) A study of visual perseveration. Journal of Neurology, Neurosurgery and Psychiatry, 26, 468-475. KRAUSE F (1924) Die Sehbahnen in chirurgischer Beziehung und die faradische Reizung des Sehzentrums. Klinische Wochenschrift, 3, 1260-1265. LANCE J W (1976) Simple formed hallucinations confined to the area of a specific visual field defect. Brain, 99, 719-734. MCAULEY D L, Ross RUSSELL R W (1979) Correlation of CAT scan and visual field defects in vascular lesions of the posterior visual pathways. Journal of Neurology, Neurosurgery and Psychiatry, 42, 298-311. MOONEY A J, CAREY P, RYAN M, BOFIN P (1965) Parasagittal parieto-occipital meningioma with visual hallucinations. American Journal of Ophthalmology, 59, 197-205. PENFIELD W, PEROT P (1963) The brain's record of auditory and visual experience. Brain, 86, 595-696. RICHARDS W (1971) The fortification illusions of migraines. Scientific American, 224, 89-96. SCHEIN S J, MARROCCO R T, MONASTERIO F M de (1982) Is there a high concentration of colourselective cells in area V4 of monkey visual cortex? Journal of Neurophysiology, 47, 193-213. TEUBER H L, BATTERSBY S W, BENDER M B (1960) Visual Field Defects after Penetrating Missile Wounds of the Brain. Cambridge, Mass.: Harvard University Press. WESTPHAL K (1881) Ziir Frage der Lokalisation der unilateralen Convulsionen und Hemianopsie bedingten Hirnerkrankungen. Charite-Annalen, 6, 342-365. WILBRAND H, SAENGER A (1917) Die homonyme Hemianopsie. Wiesbaden: J. F. Bergmann Verlag. ZEKJ S M (1973) Colour coding in rhesus monkey prestriate cortex. Brain Research, Amsterdam, 53, 422-427. (Received November 30, 1982. Revised May 17, 1983)