smooth pursuit gaze opposite from the lesion cleared within weeks. It appears that an acute, reversible loss of pursuit gaze contralateral to the lesion can occur following ablative lesions of the occipitoparietal region, and that a more chronic and partial defect in pursuit gaze ipsilateral to the lesion may persist. The acuteness and reversibility of this contralateral pursuit gaze paresis is comparable with the similarly short-lived loss of contralateral saccadic gaze occurring after a more anterior and global destruction of the frontal lobes in the middle cerebral artery distribution.3 It seems that ipsilateral and contra- lateral, conjugate horizontal eye devi¬ ation can arise from either posterior hemisphere. The gaze functions of the posterior hemispheres under normal conditions have yet to be deter¬ mined. References 1. Gay AJ, Newman NM, Keltner JL, et al: Eye Movement Disorders. St Louis, CV Mosby Co, 1974. 2. Sharpe JA, Deck JHN: Destruction of the internal sagittal stratum and normal smooth pursuit. Ann Neurol 1978;4:473-476. 3. Cogan D: Neurology of the Ocular Muscles, ed 2. Springfield, Ill, Charles C Thomas Publisher, 1956. 4. Daroff R, Hoyt WF: Supranuclear disorders of ocular control systems in man, in Bach-y-Rita P, Collins C (eds): Control of Eye Movements. New York, Academic Press Inc, 1971, pp 175\x=req-\ 235. 5. Rieck NW: Motor responses from the macaque frontal lobe. J Comp Neurol 1959; 112:203-229. 6. Wagmen IH, Krieger HP, Bender MB: Eye novements elicited by surface and depth stimulation of the occipital lobe of macaque mulatta. J Comp Neurol 1958;109:169-193. 7. Jampel RS: Convergence, divergence, pupillary reactions and accommodation of the eyes from faradic stimulation of the macaque brain. J Comp Neurol 1960;115:371-399. 8. Robillard A, Saint-Hilaire J-M, Mercier M, et al: The localizing and lateralizing value of adversive epileptic seizures. Neurology 1983; 33:1241-1242. 9. Baloh RW, Yee RD, Honrubia V: Optokinetic nystagmus and parietal lobe lesions. Ann Neurol 1980;7:269-276. Right Hemifield Alexia Without Hemianopia Alexandre Castro-Caldas, MD, Vasco Salgado, MD \s=b\ We studied a 54-year-old man who suffered a stroke in the visual-association areas of the left hemisphere. He showed no right-sided visual field defect and results of a neuropsychologic examination were normal except for his performance in several tasks presented to the right visual field. Naming of drawings was normal in both hemifields but reading of letters, words, digits, and arithmetic symbols was defective in the right field. We believe that the lesion in the visual association areas of the left hemisphere that did not interrupt interhemispheric connections disconnected the primary visual areas from temporoparietal areas of the left hemisphere, affecting reading. (Arch Neurol 1984;41:84-87) Although pure alexia is generally "^ accompanied by right-sided vi¬ sual field defects, this is not a crucial feature of the syndrome, but rather a neighboring sign consequent to the extension of the lesion to the calcarine cortex or optic radiations. Pure alexia can be the result of lesions that disconnect the pathways from both occipital lobes to the parietotemporal association areas respon¬ sible for visuoverbal association Accepted for publication March 7, 1983. From the Language Research Laboratory, Centro de Estudos Egas Moniz (Instituto Nacional de Investiga\l=c;\\l=a'\oCientifica), Lisbon. Reprint requests to Language Research Laboratory, Centro de Estudos Egas Moniz, Hospital de Santa Maria, 1600 Lisboa, Portugal (Dr Castro-Caldas). tasks.13 Such lesions are located in the left occipital region. Alexia can also be confined to the left visual field in callosotomized patients because vi¬ sual information cannot travel from the right occipital lobe to the domi¬ nant parietotemporal association areas.4 Recent evidence supports the notion that specific regions of the visual association cortex may be specialized in the processing of specific types of stimuli or tasks. Strategically located lesions in peristriate regions can inde¬ pendently affect functions such as reading, visual object naming, color perception and face recognition.5"7 Such lesions can spare the optic radia¬ tions, primary calcarine cortex, and even callosal outflow. The resulting syndromes can thus be confined to an otherwise intact visual hemifield. REPORT OF A CASE A 54-year-old right-handed office clerk was referred to our laboratory because of reading difficulties. While at work one year before admission, he suddenly com¬ plained of dizziness, headache, and nausea and noted that he could not see in the right half of his visual field. His visual difficul¬ ties improved within a few days. It is unclear if during that period he was able to read, because he apparently did not attempt to do so. On returning to work he was able to read but did so slowly because he missed words and letters and had to return to the beginning of sentences to grasp the meaning of the material. These difficulties improved during the following year but, based on his testimony, he never regained his previous reading ability. He Downloaded From: http://archneur.jamanetwork.com/ by a Monash University Library User on 06/18/2015 Fig 1.—Visual field of left eye. never complained of other neurologic symptoms. Twenty years earlier he had sustained a traumatic lesion of the right eye that ren¬ dered him completely blind on that side. Results of a neurologic examination were normal, including examination of the visual fields (Fig 1). A computed tomographic (CT) scan showed an area of hypodensity in the mesi¬ al and inferior left occipital lobe extending deeply into the white matter. It appeared to spare the calcarine cortex and the for¬ ceps major (Fig 2). As assessed by the standard neuropsychological battery of the Language Research Laboratory,89 there was no evi¬ dence of aphasia or apraxia. Construction¬ al ability, memory, and calculation were normal. The patient could read letters, words, and text without errors, although he complained of some difficulty in reading the newspaper, because his reading was slower than before. Spontaneous writing, copying, and writing to dictation were nor- ing tasks: (1) naming of colors was defec¬ tive in both hemifields but more disturbed three (7%) of the right (12 [28%] correct answers in a total of 42 stimula¬ tions); (2) naming of single letters was better when presented to the left field (26 [65%] 14 [35%] in a total of 40 presenta¬ tions); (3) reading of two-letter words was impossible in the right field and six (50% ) were correctly read in the left field (in a total of 12 presentations); (4) naming of digits was also better in the left field (ten [63%] two [12%] in a total of 16 stimula¬ tions); and (5) naming of arithmetic sym¬ bols was normal in the left field and impos¬ sible in the right (in a total of ten expo¬ sures). When the patient failed to name correctly the presented stimuli he was asked verbally to choose the items he did not name from a multiple choice card on which all of the stimuli were represented. He was able to match 21 of the 30 colors he could not name when stimuli were pre¬ sented in the left field and 24 of the 39 he could not name when stimuli were present¬ ed in the right field. He was unable to match letters, words, digits, or arithmetic symbols he could not name (results were thus similar both for naming and match¬ ing tasks). When meaningless drawings were presented in each hemifield he could easily select the correct items from a mul¬ tiple choice card. Finally, pairs of different and similar stimuli were presented simultaneously in both hemifields (eg, red in the right field and green in the left or green in both hemifields). The patient was asked to state if the stimuli presented were similar or different. Performance was normal for col¬ ors and nonverbal stimuli and a chance score was obtained when pairs of letters on Fig 2.—Computed tomographic scan showing lesion on left occipital lobe sparing primary visual were presented. cortex and corpus callosum. COMMENT Based Results of Tachistoscopic Tests Field, No. Correct/ No. Presented (%) Task Left Right Naming of drawings Naming of colors Naming of single 27/36(75) 12/42 (28) 21/36 (58) 26/40 (65) 14/40 (35) 6/12 (50) 0/12 (0) 10/16 (63) 2/16(12) 7.20, P< .01 8.00, < .005 8.53, P< .005 10/10 (100) 0/10 (0) 20.00, P< .005 letters Reading of words Naming of digits Naming of arithmetic symbols mal and the patient felt that he performed as well as ever. Naming of digits, numbers, arithmetic symbols, geometric shapes, col¬ ors, familiar faces, European traffic signs, and photographs of city landmarks was performed accurately. His score was nor¬ mal in the face recognition test10 and he had no abnormalities in the Ishihara test." He then performed several tasks during a tachistoscopic examination of both hemifields. Stimuli were presented in a tachistoscope at an exposure of 100 ms. Random¬ ized stimuli were given 2.5° to the right 3/42 (7) on the several available reports on pure alexia it is possible to 2.25, NS Ì.57, P< .025 and to the left of the fixation point. Stimuli presented to the left field were always also presented to the right. The same series of stimuli were presented several times in different days. Performances were consis¬ tent in each trial. Results are summarized in the Table. For each task the total num¬ ber of correct answers in all trials was considered. The patient correctly named 27 (75% ) of the drawings presented to the left field and 21 (58% ) of those presented to the right in a total of 36 stimulations, which can be considered a similar performance. Differences were obtained for the remain- postulate some of the areas and path¬ ways involved in reading ability. Visual information arriving at the primary visual cortex proceeds to peristriate association areas in each hemisphere. From there it travels to the left parietotemporal cortex, from the left occipital lobe through the white matter underlying the occipito¬ parietal regions, and from the right occipital lobe through the inferior splenium of the corpus callosum. Three mechanisms have been postu¬ lated by Geschwind12 for the right occipital lobe-inferior splenium path¬ way: (1) the pathway proceeds from the right Brodmann's area 17 to the right visual association areas, from there to the right angular gyrus, and finally across the corpus callosum to the left angular gyrus (Fig 3, A); (2) the pathway runs from the right Brodmann's area 17 to the right asso¬ ciation areas, then crosses the callo¬ sum to the left visual association Downloaded From: http://archneur.jamanetwork.com/ by a Monash University Library User on 06/18/2015 6 3J < 6 3J«- so 3 3M •^ 5 |ff]a 7*b a I5l a Fig 3.—Schematic representation of mechanism of connections necessary for reading. Left, Four possible mechanisms. Right, Interpretation of present case according to each mechanism. 1 indicates left visual primary cortex; 2, left visual association cortex; 3, left parietotemporal cortex; 4, right visual primary cortex; 5, right visual association cortex; and 6, right parietotemporal cortex, a, b, and c indicate possible localization of damage. visual association cortex and the left parietotemporal cortex could have been damaged, with preservation of the right-left connections (Fig 3, c) but that would not explain the patient's inability to match letters presented to the right visual field. If the lesion spared the mechanisms of visual information processing in the left occipital lobe, the patient, although unable to perform visuoverbal associations due to a disconnection between the left occipital cortex and the language areas, should still have been able to perform those associa¬ tions without using language. He should thus have been able to match all material presented to the right hemifield. If the second mechanism is consid¬ ered (Fig 3), the lesion would be responsible for the disconnection between primary and association visual cortices, otherwise all the information arriving from the right hemisphere would not reach the left parietotemporal areas and the patient would have shown a typical pure alex¬ ia. However, the lesion seemed to be too large to be responsible for such a strategic localization. For the last two mechanisms the explanation is similar to that used for the first: the lesion damaged the left visual association cortex or its connec¬ tions with the primary visual cortex (Fig 3, C and D, and a and b). Discon¬ nection of the visual association cor¬ tex from the parietotemporal cortex (Fig 3, c) is insufficient to explain the clinical picture, but could also be present. areas, and finally runs forward to the left angular gyrus (Fig 3, B); and (3) both pathways are used (Fig 3, C). A possible fourth mechanism, based on a model of nonhomotopic interhemispherical connections, would call for a pathway running from the right Brod¬ mann's area 17 to the right associa¬ tion areas, then crossing the callosum directly to the left angular gyrus (Fig 3, D). The most prominent clinical fea¬ tures of the present case were: (1) absence of abnormal findings in an open-field neuropsychologic examina¬ tion; (2) alexia for letters, words, dig¬ its, and arithmetic symbols presented in the right visual field (the defect was present in naming and matching tasks); and (3) anomia for colors pre¬ sented in the left and in the right visual fields (although less evident for colors presented in the left field) (col¬ or-matching tasks were better per¬ formed than color naming). The CT scan showed a lesion in the left occipital lobe that was damaging visual association areas and the underlying white matter. The primary visual cortex, the corpus callosum, and the forceps major seemed to be spared. Considering the first mechanism proposed (Fig 3, A), the lesion in this patient could have damaged the left visual association cortex (Fig 3, a) or disconnected the primary and associa¬ tion visual cortexes (Fig 3, b). In either case, the cascade processing of visual information in the left occipital lobe was stopped, preventing the decodification of visual symbols pre¬ sented to the right visual hemifield. However, right-left connections re¬ mained operative, allowing reading in the left field. The information pro¬ cessed in the right occipital lobe could be transported to the left parietotem¬ poral cortex through the corpus callo¬ sum. The connections between the left Downloaded From: http://archneur.jamanetwork.com/ by a Monash University Library User on 06/18/2015 We can conclude that all mecha¬ nisms except for the second (Fig 3, B) which seems improbable, are possible and that a lesion in the visual associa¬ tion cortex or in its connections with the primary cortex can cause an alexia confined to the right visual field. The patient showed more intense defects in color naming than in the naming of letters in tachistoscopic presentation, although with the same assymmetry of performance between the two hemifields. This can be explained by the time required for color detection, which has proved to be longer even in normal subjects.13 However, he performed better in matching tasks of colors than in matching tasks of symbolic visual material. It is thus probable that he grasped information that was enough for nonverbal tasks (in particular those of matching) but insufficient for verbal ones. This means that his left occipital lobe could only treat color information up to a certain point, that is, for tasks that did not require a verbal intervention. This study was supported in part by a grant from Fundaçâo Calouste Gulbenkian. A. R. Damasio, MD, provided helpful sugges¬ tions. References 1. D\l=e'\j\l=e'\rineJ: Contribution \l=a'\l'etude anatomo-$ pathologique e clinique des diff\l=e'\rentsvariet\l=e'\sde c\l=e'\cit\l=e'\verbale. Mem Soc Biol 1892;4:61-90. 2. Geschwind N, Fusillo M: Color-naming defects in association with alexia. Arch Neurol 1966;15:137-146. 3. Greenblatt SH: Subangular alexia without agraphia or hemianopsia. Brain Lang 1976;2:229\x=req-\ 245. 4. Gazzaniga MS, Bogen JE, Sperry RW: Observations on visual perception after disconnexion of the cerebral hemispheres in man. Brain 1965;88:221-236. 5. Damasio AR: Notes on the anatomical basis of pure alexia and of color naming, Sarno MT, Hook O (eds): in Aphasia: Assessment and Treatment. Stockholm, Almqvist & Wiksell International, 1980, pp 126-131. 6. Damasio AR, Yamada T, Damasio H, et al: Central achromatopsia: Behavioral, anatomic and physiologic aspects. Neurology 1980;30:1064\x=req-\ 1071. 7. Dam\l=a'\sioAR, Damasio H, Van Hoesen GW: Prosopagnosia: Anatomic basis and behavioral mechanisms. Neurology 1982;32:331-341. 8. Damasio AR: Neurologia da Linguagem. Lisbon, Buchholz, 1973. 9. Castro-Caldas A: Diagn\l=o'\sticoe evolu\l=c;\\l=a~\odas afasias de causa vascular, thesis, Lisbon, 1979. 10. Benton AL, Van Allen MW: Impairment in facial recognition in parents with cerebral disease. Cortex 1968;4:344-358. S: Tests for Colour-Blindness. Tokyo, Kanehara Shuppan Co Ltd, 1971. 12. Geschwind N: Disconnection syndromes in animal and man: Part II. Brain 1965;88:585-644. 13. Hyman B, Damasio AR: Words and colors are recognized differently in the upper and lower visual fields. Neurology 1982;32(suppl):126. 11. Ishiara Transient Oculomotor Synkinesis in Temporal Arteritis Patrick A. Sibony, MD, Simmons Lessell, MD \s=b\ In a patient with temporal arteritis, ophthalmoplegia developed associated with a transient oculomotor synkinesis or aberrant regeneration. The synkinesis suggests that the ophthalmoplegia was neurogenic rather than myogenic. Based on published experiments, it appears that transient synkinesis is compatible with the hypothesis that aberrant regeneration of axons in the third nerve is the cause of oculomotor synkinesis. (Arch Neurol 1984;41:87-88) patient with a pupil-sparing third-nerve palsy due to temporal arteritis subsequently had a transient anomaly of ocular motility, oculomo¬ tor synkinesis or aberrant regenera¬ tion, unique in this disease. The site of the lesion responsible for ophthalmo¬ plegia in temporal arteritis remains somewhat controversial, but the occurrence of synkinesis following a third-nerve palsy suggests that oph¬ thalmoplegia, in this case, was neuro¬ genic. REPORT OF A CASE In late November 1980, a 63-year-old had bifrontal pain that radiated to the malar regions and upper gingivae. The man Accepted for publication April 30, 1983. From the Division of Ophthalmology, Department of Neurology, State University of New York at Stony Brook (Dr Sibony); and the Department of Ophthalmology, Neurology, and Anatomy, Boston University School of Medicine (Dr Lessell). Reprint requests to Department of Neurology, State University of New York at Stony Brook, Health Sciences Center, Stony Brook, NY 11794 (Dr Sibony). pain was not aggravated by chewing or talking. He also had such severe scalp tenderness that contact with a pillow was intolerable. Over the ensuing weeks he lost 9 kg but had no fever, muscle aches, or visual impairment. His pain then became localized above the right eye and diplopia and ptosis of the right upper lid developed. On admission to a hospital, a pupil-sparing right-sided third-nerve paralysis was rec¬ ognized, but the results of the remainder of his examination were unremarkable. Com¬ puted tomographic scan of the head, orbit¬ al venogram, and right carotid arteriogram were normal. The right superficial temporal artery was opacified on the arteriogram and appeared narrowed, but it was partially obscured by branches of the internal carotid artery. Fasting blood glu¬ cose levels were consistently elevated; twohour postprandial blood glucose level was normal. Five Westergren ESRs ranged from 52 to 87 mm/hr. He was discharged with a diagnosis of diabetic ophthalmople¬ gia. Persistent severe pain requiring nar¬ cotic analgesics prompted referral. Examination on Jan 23, 1981, showed that his acuity was correctable to 20/15 OU. He missed all but two of the 20 plates in the Ishihara series with the right eye and made no errors with the left. Exophthalmometry was 14 mm OD and 13 mm OS with normal orbital resilience. There was moderate ptosis of the right upper lid, and the lid elevated 2 mm on adduction and on downward gaze. He had impaired eleva¬ tion, depression, and adduction of the right eye. Abduction and intorsion were normal. Pupillary size and motility were unre¬ markable. Visual field examination re¬ vealed a 2° central scotoma in the right eye to the 2-mm white object as 2 m. The results of the remainder of the neurologic and ophthalmologic examinations were unremarkable. Shortly after being seen, the patient's ptosis and diplopia spontaneously im- proved so that by March 6,1981, ocular and lid motility seemed normal both to the patient and to his neurologist. The intensi¬ ty of the pain diminished; however, he noticed that his scalp vessels had enlarged. A repeated ESR of 117 mm/hr led to a right-sided temporal artery biopsy, which immediately and permanently eliminated his pain. Biopsy showed giant cell arteritis. Treatment started on March 9, 1981, with oral prednisone, rapidly normalized his ESR. Repeated examination on March 20, 1981, showed that his visual acuity was 20/15 OU with normal color vision, full visual fields, and normal motility. Neither disc pallor nor thinning of the retinal nerve fiber layer could be discerned in the fundi. COMMENT Estimates of the prevalence of ophthalmoplegia in patients with tempo¬ ral arteritis vary from 10% to 50%.''3 Fisher2 has observed that the ophthalmoplegia tends to be manifested by ptosis, "partial third-nerve involve¬ ment," or, less commonly, abduction paresis. As in our patient, recovery tends to be complete and may occur without treatment. Barricks et al4 have reported autopsy findings in a patient with ophthalmoplegia conse¬ quent to temporal arteritis. The oph¬ thalmoplegia was due to muscle isch¬ emia. In addition, there are some clin¬ ical observations that suggest a myo¬ pathie basis for some cases of oph¬ thalmoplegia in temporal arteritis. Wagener and Hollenhorst1 noted daily variation in the muscles involved and severity of the ophthalmoplegia, which seemed more compatible with a myogenic rather than a neurogenic process. In some cases of ophthalmo- Downloaded From: http://archneur.jamanetwork.com/ by a Monash University Library User on 06/18/2015