Journal of Neuro-Ophthalmology, December 1994, Volume 14, Issue 4

The anterior visual pathways.

Journal of Neuro- Ophthalmology 14( 4): 234- 249, 2994. © 1994 Raven Press, Ltd., New York Part Two Annual Review in Neuro- Ophthalmology The Anterior Visual Pathways Alfredo A. Sadun, M. D., Ph. D., and Jung Dao Optic Nerve The optic nerve has always been the focus of a great deal of attention in neuro- ophthalmology. In the last two years there have been several articles presenting results of the Optic Neuritis Treatment Trial ( ONTT). About 450 patients at 15 centers were randomized to one of the three groups to receive oral prednisone, intravenously adminis­tered methylprednisolone followed by oral pred­nisone, or placebo ( 126,127). In a landmark study, published in the New England Journal of Medicine, it was found that at 6 months follow- up, intravenous methylprednisolone followed by oral prednisone led to a better recovery of vision in patients with optic neuritis than did placebo or oral prednisone ( 126). Even more remarkably, it was discovered that oral prednisone alone increased the risk of new episodes of optic neuritis, so its use should be discouraged ( 88,128). At 1 year follow- up, the dif­ference between placebo- treated patients and those on intravenous methylprednisolone had vanished ( 129). Hence the only value for intrave­nous treatment would have been to hasten visual recovery ( the rate of improvement was faster for patients receiving intravenous treatment, but the overall improvement was not greater). Most of the patients entering the Optic Neuritis Treatment Trial had their baseline visual fields carefully studied ( 130). Central or centrocecal sco­tomas were found in only about 8% of the eyes of patients with optic neuritis as demonstrated on the From the Department of Ophthalmology, Doheny Eye Insti­tute, University of Southern California School of Medicine, Los Angeles, California, U. S. A. Address correspondence and reprint requests to Dr. Alfredo A. Sadun, Estelle Doheny Eye Institute, 1450 San Pablo Street, Los Angeles, CA 90033, U. S. A. Humphrey visual fields analyzer. Approximately 48% had diffuse visual field loss and 20% had nerve- fiber bundle- type defects. Of further interest was the fact that the fellow uninvolved eye in over two- thirds of the cases had minimal visual field defects as well ( 130,131). In rebuttal, Pollack ( 132) criticized these findings largely because the visual fields were done using Humphrey automated pe­rimetry. Pollack ( 132) argued that properly per­formed visual fields would have greater diagnostic value and demonstrated more defects. Keltner and colleagues ( 133) responded to this criticism that au­tomated visual fields are now the " gold standard" and therefore, for practical considerations, the vi­sual field test most likely to occur in common us­age. In addition to visual field defects, the fellow eyes of patients with optic neuritis demonstrated color vision abnormalities in 22%, and contrast sensitiv­ity defects in 15%. Only 14% had more than one line of decreased visual acuity ( 134). The ONTT also demonstrated that routine diagnostic studies, including ANA testing, were generally not helpful ( 135). It has been suggested that steroids are particu­larly likely to be beneficial when given to patients with extremely severe visual loss due to optic neu­ritis. It has been hypothesized that steroids might reduce the extreme edema that accompanies de­my elination reducing relative ischemia. As men­tioned previously, this issue was addressed di­rectly by the Optic Neuritis Treatment Trial, and it was found that prednisone treatment did not re­duce the number of poor outcomes ( 136). In general, the early ONTT reports led to the following conclusions: 1. Diagnostic studies, including magnetic reso­nance imaging ( MRI), are not necessary. 234 ANTERIOR VISUAL PATHWAYS 235 2. Oral prednisone was definitely contraindi-cated. 3. Intravenous methylprednisolone followed by oral prednisone was probably of slight value ( 126- 128,135,137). Additionally, in an editorial that followed the New England Journal of Medicine article, Lessell ( 138) ad­dressed the association between optic neuritis and multiple sclerosis. Noting that most patients with optic neuritis have the same high- intensity signals on MRI as do multiple sclerosis patients, and that the clinical characteristics of optic neuritis whether associated or not associated with multiple sclerosis are the same, he concluded that optic neuritis is simply an attack of multiple sclerosis, and the re­sults of the Optic Neuritis Treatment Trial might thus be extrapolated to issues of management for multiple sclerosis ( 138). Most recently, another article looking at a longer term follow- up of patients in the ONTT presented bombshell findings. The patients treated with in­travenous methylprednisolone for optic neuritis had a reduced incidence of multiple sclerosis of more than 50%(!) ( 139). At 2 years follow- up it was found that symptoms of definite multiple sclerosis in patients who presented with optic neuritis could be reduced from approximately 17% to less than 8% by the administration of intravenous methyl­prednisolone. The huge protective effect of intra­venous methylprednisolone also led to a reassess­ment of the value of MRI scans ( 137,140). It is now believed that if an MRI shows two or more uniden­tified bright objects ( plaques) 3- mm in diameter, the patient is likely to develop multiple sclerosis, or at least have a recurrence of optic neuritis. Thus this cohort should receive intravenous methyl­prednisolone, since this treatment would be given not for its effect on visual outcome, but rather to " slow down" the development of multiple sclero­sis ( 139). In an editorial reviewing these data and the recommendations made by the Optic Neuritis Treatment Trial group, Savino ( 141) concluded that 1. high- dose intravenous corticosteroid side ef­fects are minimal; 2. visual deficits in the fellow eye are common; 3. the standard extensive workup for patients with optic neuritis is unnecessary except for brain MRIs to assess the patient's risk for multiple scle­rosis; 4. corticosteroids speed up the recovery but have no influence on final visual income; 5. severity of initial visual loss is a predictor of final visual income; and 6. he personally would now perform brain MRIs on all patients with optic neuritis, and when the MRI was suggestive of multiple sclerosis, he would encourage treatment with intravenous methylprednisolone regardless of the patient's vi­sual status. The downside of treatment with intravenous methylprednisolone has been addressed as well ( 142). Major side effects ( depression and acute pancreatitis) were rare and easily reversed ( 142). Further follow- up data from the ONTT study is awaited with great interest. Further surprises may await us. At this junction my take- home message is, in cases of acute optic neuritis, do an MRI and, if positive, give intravenous methylprednisolone. I present this to the patient as a window of opportunity to reduce or slow down the development of multiple sclero­sis. Why such high doses of methylprednisolone protect against multiple sclerosis remains a ques­tion for future research. An odd new treatment for multiple sclerosis has also been proposed recently. A medicinal plant, Ruta graveolens, is ascribed to contain potassium channel blockers ( 143). Nine patients with acute optic neuritis were treated with a single dose ex­tracted from this plant, and in 5 of the 9 patients the scotoma became smaller for a period of 24 to 48 hours. While it is interesting to theorize why blocking potassium might be advantageous in a demyelinative state, the lack of control subjects and other methodological problems makes it very hard to analyze this paper ( 143). In the meantime, work is continuing on an optic neuritis model in the guinea pig ( 144). Endoge­nous hydrogen peroxide produced a breakdown of the blood- brain barrier in the optic nerve of guinea pigs sensitized for experimental allergic encepha­litis ( 144). Such hydrogen peroxide might derive from a variety of inflammatory cells seen around blood vessels in the optic nerve during episodes of optic neuritis. Celesia and colleagues ( 145) presented contrast sensitivity gratings to patients with multiple scle­rosis and to controls and then analyzed the visual evoked responses by fast Fourier transform. They found that one- third of their multiple sclerosis pa­tients showed abnormalities at all spatial frequen­cies tested, and more than half had abnormalities to at least one spatial frequency. These findings corroborate the fact that many patients with mul­tiple sclerosis have had subclinical attacks of optic neuritis. In another study, 104 patients with mul­tiple sclerosis and normal visual acuity had mea­sures of visual acuity, visual evoked response ( 8), / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 236 A. A. SADUN AND J. DAO contrast sensitivity functions, brightness sense testing, and color vision testing ( 146). In their study group, 22% had a previous history of acute optic neuritis, but this group was distinguishable from the larger group only by contrast sensitivity testing. In general, most patients had normal vi­sual acuities, normal contrast sensitivities and nor­mal color vision ( 146). In the same issue of Neuro-ophthalmology is a description of 60 eyes of 50 pa­tients with optic neuritis who had been treated with high doses of methylprednisolone in a proto­col quite similar to that used in the Optic Neuritis Treatment Trial ( 147). This study, done in France, found dramatic improvement in vision in 56 of 60 eyes, and normalization of visual fields in a kinetic pattern that they described as " progressive cen­tripetal amelioration" ( 147). From Japan came a review of 14 patients with idiopathic optic neuritis in whom visual acuities were between no light perception and counting fingers ( 148). Despite having been selected for such poor vision, about three- quarters of them showed improvement of visual acuity to 20/ 40 or better; however, in about one- quarter of the eyes, the visual acuity remained below 20/ 200 ( 148). Also from Japan came a description of a 3- year- old child with bilateral optic neuritis ( papillitis) documented both clinically and with MRI ( 149). Not surpris­ingly, her vision recovered with corticosteroids. A case of bilateral visual loss and optic disk swelling in a 9- year- old child diagnosed to be optic neuritis ( papillitis) provided the basis for a review of the management of optic neuritis in children ( 150). Optic neuritis has also been described to be asso­ciated with familial Mediterranean fever ( 151), which is a poorly understood inherited disorder characterized by recurrent fevers. Two young pa­tients of Sephardic background, with an 8- to 10- year history of familial Mediterranean fever, devel­oped painless blurring of vision in the setting of normal neuroradiological studies ( 151). In both cases visual acuities returned to normal and, since both patients were in their twenties, one must wonder whether this represents isolated optic neuritis of an idiopathic or multiple sclerosis-associated nature that, by happenstance, was seen with familial Mediterranean fever. Optic neuritis in combination with parainfec­tious encephalitis was reported in two children in whom there was also increased signal in the thal­amus on T2- MRI ( 152). Biopsy of the thalamic le­sions demonstrated chronic perivascular inflam­mation with activated microglia in a nodular for­mation. Both patients improved following steroid administration ( 152), but in both cases initial at­tempts at tapering the corticosteroids led to re­lapses. Long- term corticosteroid use resulted in good final visual acuity. A 20- year- old black patient whose initial workup was normal was diagnosed as having bilateral ret­robulbar neuritis ( 153). Attempts at tapering corti­costeroids led to relapses of optic neuritis. Such a finding requires a complete systemic evaluation to exclude a collagen vascular diathesis and sarcoid­osis. Further workup demonstrated elevations of angiotensin- converting enzyme and erythrocyte sedimentation rate, leading to a diagnosis of sar­coidosis ( 153). A 52- year- old black man with sickle cell SS dis­ease had a sudden loss of vision in one eye and an amaurotic pupil. His examination and workup were entirely normal until 2 months later, when optic atrophy developed. The diagnosis is of pos­terior ischemic optic neuropathy secondary to sickle cell disease. This is largely a diagnosis of exclusion ( 154). A 40- year- old patient is described with progres­sive loss of vision over 1 month to light perception and in whom disk edema was noted ( 155). Workup was otherwise negative and she was put on corti­costeroids, which led to prompt return of her vi­sion to 20/ 40 within the next 2 weeks. However, 4 months later her vision again decreased and fur­ther neuroimaging demonstrated a lesion compat­ible with sarcoidosis in the temporal fossa ( 155). Such cases serve as regular reminders that ste­roid- responsive optic neuritis may have a specific etiology, and, when patients show reexacerbation of symptoms on steroid taper, further workup is mandated. Retrobulbar optic neuropathy may occur as an indirect consequence of Wegner's granulomatosis ( 156), that is, compression or vasculitis. In the ab­sence of any orbital disease, a 55- year- old woman developed bilateral severe optic neuropathies with small focal lesions seen on MRI in the posterior optic nerves ( 156). These findings probably repre­sent a local vasculitis. Although nasopharyngeal carcinomas often lead to ocular symptoms, it is rare that such patients present initially with visual loss. However, cases have been described in which profound visual loss was the initial presentation of nasopharyngeal car­cinoma, as the tumor had extended to the optic canal, producing isolated optic neuropathy ( 157). Radiation therapy failed to lead to any visual re­covery, but may have delayed the onset of oph­thalmoplegias. Visual recovery from radiation- induced optic neuropathy has been attributed to the use of hy- / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 ANTERIOR VISUAL PATHWAYS 237 perbaric oxygen therapy ( 158). Hyperbaric oxygen therapy was given to a 45- year- old woman who suffered severe visual field loss bilaterally followed by the visual acuity going to no light perception in one eye, 1.5 years after receiving 5,000 cGy of ra­diation therapy. Within 2 weeks of the initiation of hyperbaric oxygen therapy, the temporal hemi-anopia cleared in the less affected eye and light perception returned to the more severely affected eye. Repeat MRI studies showed decreased swell­ing and less gadolinium enhancement in the optic nerves ( 158). Two cases are reported in which embolization of the internal maxillary artery was required to con­trol severe epistaxis ( 159). Unfortunately, in both cases the ophthalmic artery was occluded. A case of a 67- year- old man who underwent bilateral neck dissection that resulted in bilateral posterior isch­emic optic neuropathy was described ( 160). Histo­pathologic analysis revealed infarction of the or­bital portions of both optic nerves, which led to the hypothesis that postoperative hypotension had caused the problem ( 160). In addition to hypoten­sion, anemia associated with surgery may be a sig­nificant risk factor for the development of posterior ischemic optic neuropathy ( 161). Amiodarone, a medication given for cardiac ar­rhythmia, may have contributed to a bilateral pro­gressive optic neuropathy seen in a 62- year- old woman ( 162). However, features of ischemia in this case may have complicated the picture ( 162). It is well known that ethambutol, like a number of other antibiotics, can produce an optic neuropa­thy. This neuropathy is felt to be reversible if caught early. This belief is addressed in a series of seven consecutive patients receiving this drug ( 163). Of the seven patients, three developed se­vere optic atrophy even after discontinuation of the drug at first visual symptoms, and only three recovered vision to the 20/ 200 or better level. The authors concluded that, given the poor reversibil­ity of vision loss following ethambutol optic neu­ropathy, this drug should be used rarely as a first-line antituberculous medication ( 163). Addition­ally, the authors suggested that visual evoked potentials are valuable in the detection and moni­toring of the optic neuropathy associated with ethambutol ocular toxicity. I find electrophysio­logic tests to rarely be more sensitive than an in-office clinical assessment. Similarly, a Japanese study of familial optic atrophy found that elec-troretinograms were of no assistance ( 164). They described two unrelated young adults who devel­oped bilateral optic atrophy with deterioration in vision. Whereas dark adaptometry was abnormal, electroretinograms were entirely normal, raising the possibility of a newly described congenital dis­order ( 164). Also from Japan is a case report of a huge glioma that involved the right optic nerve and chiasm of a 6- month- old girl for whom proptosis was the pre­senting symptom. She underwent craniotomy for resection of the lesion, and the histology con­firmed the clinical diagnosis of a pilocystic astro­cytoma, grade I ( 165). While initially her visual acuity was judged to be no light perception and light perception, respectively, by 18 months of age she had a visual acuity of 20/ 60 in the better eye and has done well subsequently ( 165). The authors were able to monitor this recovery with visual evoked potentials; however, I am concerned that difficulties in assessing visual acuities at this early age, and the natural maturation of the visual sys­tem may have confounded their interpretations. Dutton ( 166) has authored a comprehensive re­view of gliomas of the anterior visual pathway. In his very extensive review of the literature he con­cludes that this intrinsic optic nerve tumor occurs primarily in the first decade of life when it may grow rapidly, but then generally falls into a phase of stability. Hence the long- term prognosis is really not so bad, although there is an overall mortality rate of about 5%. Generally, conservative manage­ment is recommended, but in cases of complete blindness and pain or severe proptosis, surgical resection may be indicated ( 166). Four young children with neurofibromatosis type I, in whom original neuroimaging studies were negative but in whom gliomas of the anterior visual pathways eventually cropped up, have been described ( 167). While the authors did not suggest that this shows that these type of tumors can be acquired later in childhood, they do point out that initial negative neuroimaging should not be the basis for telling parents that their children will not develop visual problems in the future ( 167). In the case of neurofibromatosis type II, a 34- year- old woman presented with hearing loss and bilateral optic neuropathies ( 168). Computed tomography ( CT) scanning demonstrated what appeared to be bilateral optic nerve sheath meningiomas in addi­tion to the expected bilateral acoustic neuromas ( 168). Meningiomas of the optic nerve sheath are also reviewed by Dutton ( 169), who discusses the various modalities of therapy, including conserva­tive follow- up, surgical resection, and radiation. Optic nerve sheath meningiomas are also re­viewed from a neuroimaging point of view ( 170). In 13 patients MRI sometimes proved advanta­geous over CT scanning in characterizing optic / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 238 A. A. SADUN AND J. DAO nerve sheath meningiomas in the orbit, canal, or intracranially ( 170). A new method for delivering precise radiation therapy for unilateral optic nerve sheath meningiomas has also been described ( 171). A 66- year- old man had an initial diagnosis of normal pressure glaucoma but was subsequently found to have an optic neuropathy due to com­pression from a meningioma ( 172). This confusion was due to the nasal step seen on his visual fields and the classical cupping and loss of optic disk rim seen on funduscopy. We are reminded that pa­tients with low tension ( normal pressure) glau­coma should undergo neuro- ophthalmological workup. This article is followed by a remarkable editorial by J. L. Smith ( 173). He allows us to eavesdrop on some of the reviewers' comments on this paper, such as that this patient did not, by the criteria of Trobe ( 1980), meet the definition of glau­coma. Because of this serious objection, the paper was submitted to a glaucoma expert who agreed that the compressive lesion did, indeed, closely mimic normal tension glaucoma ( 173). Further clues that this was not glaucoma may have been that ( a) the nasal field defect was inferior ( not su­perior, as usual), and ( b) the patient reported blurred vision each morning ( 173). Gaze- induced amaurosis is a well- known phe­nomenon usually reflecting intermittent compres­sion from an orbital mass. A case of gaze- induced amaurosis was described in which color Doppler imaging demonstrated a decrease in flow through the central retinal artery corresponding to those times and ocular position at which the patient could not see ( 174). The mass was an orbital varix, which was subsequently removed, after which the patient no longer noted any visual dysfunction and color Doppler imaging showed normal flow through the central retinal artery ( 174). A metastatic lesion to the optic nerve was de­scribed by Mansour and colleagues ( 175) in a 41- year- patient with a 2- week history of progressive visual loss. It was unusual in that no other cranial nerves were involved and there was no other evi­dence of an orbital process. A paraneoplastic syn­drome producing an optic neuropathy was also de­scribed in a patient who, in addition to having pro­gressive bilateral loss of vision, was noted to have ataxia and downbeat nystagmus ( 176). Although MRI of the brain was normal, the cerebrospinal fluid protein was found to be elevated. Ultimately, the diagnosis of small cell carcinoma was made on biopsy of a pulmonary lymph node. These cells were strongly positive for neuron- specific enolase, which is also found in astrocytes, possibly leading to this cross- specificity ( 176). Horton and colleagues ( 177) described MRI of the optic nerve in a case of acute myelocytic leu­kemia. This was in a 58- year- old man who com­plained of visual loss in one eye and in whom fun­dus examination showed disk edema and central retinal artery occlusion. On MRI, a bright signal filled a very extended subarachnoid space, indicat­ing leukemic infiltrate ( 177). A lymphomatous in­filtrate of the optic nerve was described in a 21- year- old man who had a diagnosis of Hodgkin's disease for 3 years ( 178). He presented with visual loss and disk edema, and on MRI scannings showed diffuse enlargement of the optic nerve. Combined radiation therapy and corticosteroids resolved the disk edema and led to a marked im­provement in visual acuity ( 178). An 84- year- old patient presented with severe visual loss and pro-ptosis and was found to have a small cell malig­nant lymphoma ( 179). Local radiation therapy led to resolution of both proptosis and improvement in vision to 20/ 40 ( 179). A 40- year- old patient with progressive loss of vision was found by biopsy to have a plasma cell granuloma ( 180). Corticosteroid therapy led to tumor shrinkage and visual im­provement ( 180). From Belgium came a description of a sphenoid sinus mucocele that did not cause the usual proptosis; rather, the patient developed sudden total blindness in one eye and then, over the next 5 days, in the other eye ( 181). Following surgery there was some recovery of vision in the second eye. From Japan came two papers describing new psychophysical measures of optic nerve function. Contrast sensitivity testing was done using both stationary and motion ( drifting vertical stripes) paradigms, and was found to be a sensitive mea­sure of the impairments associated with optic neu­ritis and other optic neuropathies ( 182). Temporal modulation transfer functions of vision in patients with optic neuritis were also investigated ( 182). In patients with optic neuritis, an abnormality could best be detected by estimating the first and second harmonic components ( 183). The quantified rela­tive afferent pupillary defect was correlated to vi­sual fields measured by both static and kinetic pe­rimetry in 137 patients with a variety of optic neu­ropathies ( 184). In some categories of optic neuropathy the correlation was strong, but in other categories, such as optic neuritis, the corre­lation was poor ( 184). Cryptococcal meningitis is becoming more com­mon, largely because of the increase in AIDS. Sud­den bilateral visual loss is a frequent occurrence with cryptococcal meningitis ( 185). A postmortem histopathologic study of the visual pathways from / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 ANTERIOR VISUAL PATHWAYS 239 such a patient with cryptococcal meningitis re­vealed focal areas of necrosis scattered throughout both optic nerves, as well as heavy infiltration of the meninges by the fungus ( 185). In a clinical study of 80 AIDS patients with cryptococcus infec­tions, it was found that 33% had papilledema, 9% had visual loss, and 9% had a sixth nerve palsy ( 186). The presence of papilledema did not corre­late with longevity in these patients. However, it is not always possible to characterize the nature of the optic neuropathies seen in HIV- positive men ( 187). Two HIV- positive patients were described as having bilateral optic neuropathies that could not be characterized. One patient received AZT and the other steroids, and both did well ( 187). While it is impossible to be sure as to what was the etiology of the visual loss or even if the treatment were related to the resolution of the problem, this paper is useful in reminding us that the prognosis is not uniformly dismal in HIV- positive patients who have an optic neuropathy. In the same journal is­sue was a description of an HIV- positive man who had subacute syphilitic meningitis that led to bilat­eral optic neuritis ( 188). Periocular corticosteroid injections may have helped salvage his vision. In a review of 78 asymptomatic HIV positive men ( 156 eyes) in which there was no retinitis, defects of color vision, and contrast sensitivity sug­gested a subclinical primary optic neuropathy ( 189). This was corroborated by a histopathological study in which optic nerves from patients with AIDS but with no retinitis were examined with a variety of special stains ( 190). A diffuse loss of ax­ons was noted. Morphometry indicated that the pattern of axonal dropout was not limited to one fiber type. Histologic studies of 66 postmortem optic nerves from normal human subjects revealed that mast cells, frequently found in other ocular elements, could also be found scattered about in the menin­ges of almost all of these specimens ( 191). It is interesting to speculate that these mast cells might mediate certain inflammatory conditions of the op­tic nerve, perhaps by the release of hydrogen per­oxide as described in an experimental model above ( 144). Histopathologic studies of the optic nerves dur­ing autopsy may provide us with useful informa­tion on other general issues. For example, Budenz and colleagues ( 192) looked at the optic nerves from 13 infants who died from acute intracranial injuries due to either blunt trauma or violent shak­ing. Optic nerve sheath hemorrhages were found in all 13 cases, but not in 6 other infants who died from sudden infant death syndrome. Metastatic le­sions of the optic nerve are often found at the level of the optic canal. In reviewing five patients with metastatic prostrate cancer, Kattah and colleagues ( 193) demonstrated compressive optic neuropa­thies clinically ( on the basis of MRIs) and later his-topathologically. These metastases were found in the epidural and subperiosteal spaces near the op­tic canal. The existence of metastases in this closed space probably explains why the visual loss was remarkably acute ( one to several days) ( 193). Ra­diotherapy and pulse intravenous methylprednis-olone was somewhat helpful in these cases. The reader is encouraged to appreciate the characteris­tic pattern of bone hypertrophy and deformity that is seen on MRI with optic nerve compression in the optic canal. The optic canal is also a subject of interest in cases of posttraumatic visual loss due to optic nerve injury. In a follow- up of their classic 1990 paper ( 194), Levin, Joseph, Rizzo, and Lessell ( 195) present further retrospective analysis of 31 cases in which transethmoidal decompression of the optic canal was performed in patients in whom head trauma led to visual loss. Interestingly, they found that younger patients showed markedly better im­provement in vision than did patients 40 or older. This seemed to be the most important variable ( as compared with preoperative visual acuity, delay to surgery, or the presence of fractures), and hence should certainly be incorporated into the long-awaited randomized trial for optic canal decom­pression for traumatic optic neuropathy. Girard and colleagues ( 196) also used a transeth­moidal approach for canal decompression but ex­tended this surgery into the sphenoid sinus in de­compressing optic nerves following indirect trauma. In 8 of 11 patients ( including 4 patients with no light perception) they were able to docu­ment marked improvement in vision following surgery ( 196). Although they had no control group, they felt that their results compared favor­ably to those in the literature, in which spontane­ous improvement of optic neuropathy following blunt head trauma is usually in the 33% range ( 197). Optic Chiasm Closed head injury can lead to a traumatic chi­asmal syndrome. Although such injury is uncom­mon, and not readily visible on neuroimaging, it needs to be considered particularly when the vi­sual field is suggestive ( 198). In one report, excel­lent MRIs demonstrated swelling of the optic chi­asm with a hemorrhage that dissected up into the / Neuro- Ophthalmol, Vol. 14, No. 4> 1994 240 A. A. SADUN AND J. DAO hypothalamus of a 17- year- old who suffered blunt head trauma ( 198). Bitemporal hemianopsias as well as evidence of diabetes insipidus triggered the appropriate workup in this case. Neetens ( 199) has published a review of a traumatic bitemporal hemi-anopia. He described how trauma to the forehead region may produce a type of splitting of the op­tic chiasm with the result of an immediate loss of vision, possible seesaw nystagmus, and some­times hypothalamic symptoms. There is an excel­lent section describing the three mechanisms by which such damage can occur to the optic chiasm ( shearing, contre- coupe, and stretching separa­tion) ( 199). Another case report describes a 42- year- old man who suddenly awoke with unilateral visual loss with visual fields suggesting a junc­tional scotoma ( 200). Neuroimaging demonstrated a hemorrhage associated with a cavernous heman­gioma ( 200). A case is described also of a 21- year-old women with progressive visual loss and what appeared to be a glioma or meningioma at the junction of the optic nerve and chiasm ( 201). The MRI showed this to be an annular enhancing le­sion and surgical excision demonstrated this to be a choristoma ( 201). Three cases are described of children, ages 4 through 8, who had presumed chiasmal gliomas ( based on MR imaging) ( 202). Without any medical or surgical intervention, all three showed improvement in visual function ( 202). Another case described is that of a 43- year-old pregnant woman who, in her third trimester, developed visual loss that was found to be second­ary to several aneurysms in the area of the optic chiasm ( 203). In this case care had to be taken to not endanger the fetus with workup or aneurys­mal clipping. Two patients are described in whom chiasmal compression occurred following transphenoidal resection due to excessive packing of fat during surgery ( 204). In one case the excessive fat was removed, and the patient showed prompt recov­ery of vision. In the other, no attempt was made to remove this excess fat and the visual loss persisted ( 204). Another case is that of a patient in whom a pituitary macroadenoma was followed by serial magnetic resonance imaging scans to the point of spontaneous involution ( 205). This led to an empty sella syndrome, and documented the long-suspected relationship between empty sellar syn­drome and infarction of a pituitary adenoma ( 205). In cases of optic chiasm dysfunction, the appro­priate visual field test and interpretation often points to the appropriate neuroimaging. Martin- Boglind ( 206) described the use of visual fields and computer- assisted interpretations in over 300 pa­tients, about half of whom had known chiasmal or perichiasmal lesions. There was approximately a 30% false negative rate, with only a 1% false pos­itive rate, suggesting that this technique is useful, but not of sufficient reliability to forego the usual neuroimaging in suspicious cases. Chiasmal and perichiasmal lesions can mimic low- tension glaucoma. In a review of 62 patients with " classical signs of low tension glaucoma," neuroimaging demonstrated that 90% had pathol­ogy of the intracavernous carotid arteries ( 207). Even more remarkably, about half of these patients had asymmetry of the optic nerve head cups, which corresponded to the side of greater carotid artery involvement ( 207). Most neuro- ophthalmolo-gists are aware that dolichoectasias of the carotids may produce binasal field defects that sometimes look like glaucoma. The Optic Tracts and Primary Visual Nuclei Few clinical studies have explored the role of ophthalmologic disease at the level of the primary visual nuclei. The lateral geniculate nucleus is the major primary nucleus, and the destination of most of the retinal ganglion cell fibers. In a histo­pathologic study of the lateral geniculate nucleus, it was found that patients with glaucoma showed more losses in the magnocellular layers ( I and II) than in the parvocellular layers ( III- IV) ( 208). This is further corroboration of the hypotheses, de­scribed by Quigley and others ( 303), that glaucoma may have a preferential effect on the magnocellu­lar system and hence that appropriate screening tests should be performed that target the psycho­physical functions of this system. Further histopathologic work on the lateral ge­niculate nucleus in humans was done on a human brain obtained from an individual who had been noted to have strabismic amblyopia ( 100). Signifi­cant shrinkage of cells was found in both lateral geniculate nuclei corresponding to those layers in­nervated by the amblyopic eye. This was true for both magnocellular and parvocellular layers, but was more prominent in the ipsilateral lateral genic­ulate nucleus ( 209). Interestingly, this corresponds closely to findings from the monkey model of stra­bismic amblyopia and suggests that structural changes occur at the lateral geniculate nucleus, which is more than a mere relay station to the vi­sual cortex. Remarkably, there was a report that levodopa administration can improve visual acu­ities and reduce the size of scotomas in human amblyopia ( 210). These authors extended their psychophysical measures of the effect of levodopa / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 ANTERIOR VISUAL PATHWAYS 241 in a double- masked crossover study and showed that over two- thirds of the patients improved sig­nificantly after 1 week of such therapy. It is unclear why the patients got better, why their improve­ment persisted after discontinuation of therapy, or at what anatomic site this may have had an effect. Three patients are described in whom arteriove­nous malformations caused damage to the optic radiations ( 211). The band atrophy noted in the contralateral optic nerve suggested retrograde transsynaptic degeneration across the lateral ge­niculate nucleus. However, high quality neuroim-aging revealed that the deep venous drainage from each arteriovenous malformation would directly involve the lateral geniculate nucleus, and hence damage the optic tract directly, not requiring trans­synaptic degeneration as a mechanism ( 211). The contribution of another primary visual nu­cleus was described in a report from Austria of a patient in whom a hemorrhage that damaged the right superior colliculus and right pulvinar of the thalamus ( as demonstrated by MRI) led to disas­sociated unilateral convergence paralysis ( 212). Both the superior colliculus and the pulvinar are known, on the basis of animal experimental mod­els, to control ocular motility ( see section to follow on anatomical pathways. PARALLEL PROCESSING IN THE ANTERIOR VISUAL SYSTEM Parallel processing refers to separate channels through which information can be independently processed. Vision scientists have considered par­allel processing in the visual system of a variety of animals for many years. Yet, ophthalmologists have tended to see the human visual system as serial processing. Many textbooks, even in n e u r ­ophthalmology, emphasize that visual information proceeds from the retina to the lateral geniculate nucleus only to " relay" to the visual cortex via the optic radiations. Light, entering the eye, is transduced into elec­trochemical signals by rods and cones that then input into bipolar, amacrine, and horizontal cells connecting to the retinal ganglion cells. Bishop ( 213) first suggested in 1933 that there are three types of retinal ganglion cells with different elec-trophysical properties and conduction velocities. Different morphological schemes of classifying ret­inal ganglion cells were variably proposed leading to Polyak's ( 214) classification of four types of pri­mate retinal ganglion cells. In 1966, Enroth- Cugell and Robson ( 215) described X and Y cell types and in 1974 Boycott and Wassle ( 216) provided corre­sponding morphologic classes of retinal ganglion cells in the cat. More recently, these anatomic and physiological classes of retinal ganglion cells were described in the primate retina ( 217). Most gener­ally, the anterior visual pathways of primates can be considered to be composed of three distinct par­allel pathways of retinofugal projections to subcor­tical primary visual nuclei ( 218,219). About 80% of all retinal ganglion cells are of moderate size and have color- opponent physiologic responses and project to the parvicellular layers of the dorsal lat­eral geniculate nucleus; these cells are often re­ferred to as the P ( or P- beta) ganglion cells. Inter­ruption of this P pathway leads to deficiencies in color vision, visual acuity, and contrast sensitivity at high spatial and low temporal frequencies ( 220). In contrast, the M pathway seems to be primarily involved with color ignorant large cells that medi­ate information of high temporal and low spatial frequency contrast sensitivity, and probably mo­tion ( 221). The retinal ganglion cells of the M cell pathway have large visual field areas, fast conduc­tion speeds and project to the two magnocellular layers of the retinal ganglion dorsal lateral genicu­late nucleus; these cells have also been called alpha cells. A third ( gamma) class of cells is less well understood and probably projects mainly to the superior colliculus, the pretectum and other brain­stem primary visual nuclei ( 218,219). In the last few years there has been a great em­phasis on the segregation of the M and P pathways ( 222). Human psychophysical and developmental studies have demonstrated the great significance of this segregation. For example, it has been pro­posed that a human neonate's vision is entirely subcortical and only begins to have cortical ele­ments at about two months postnatally. Thereaf­ter, the M and P cell pathways may underlie fur­ther visual development ( 223). The implication is that the basic behavior of a neonate relies only upon the rapid detection of objects in peripheral field such as might be processed by the superior colliculus and other primary visual nuclei of the brainstem that receive input from the primitive gamma class of retinal ganglion cells. Thereafter, cortical integration of spatial representation sys­tems for color and form take place, suggesting in­volvement of the P cell system. Finally, the M cell system contributions of relative position, motion, and depth come in. Hence it is suggested that hu­man development goes from gamma class gan­glion cells to the P cell system to the M cell system ( 223). At the level of the visual cortex, the anatomic segregation of the M and P pathways remains / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 242 A. A. SADUN AND J. DAO largely complete even to the extent that there re­mains separation of axon terminal arboration fields and dendritic fields ( 221). This anatomic segrega­tion begins, of course, in the retina with evidence that there is a distinction between the topography of the M and P cell ganglion cells. M cells, identi­fied by neurofibrillar staining, probably constitute 6% to 10% of the retinal ganglion cells except for the nasal horizontal meridian where they repre­sent up to 20% of the retinal ganglion cells ( 218). This is interesting because this nasal horizontal meridian is represented by the monocular visual field of the cortex. Most of the other retinal gan­glion cells ( about 80%) are P ganglion cells, which are concentrated in the fovea but remain the dom­inant cell form across most of the rest of the retina as well ( 220). The consequence of selective damage to M or P cell ganglion cells have been measured psycho-physically in man and monkey as well. Selective destruction of P cells by a neurotoxicant led to de­creases in visual acuity in all areas of the visual field ( 220). However, there was preservation of contrast sensitivity, vernier acuity, and even shape discrimination despite this chemoablation. Similar studies have revealed that M and P ganglion cells show a clear difference in their contrast sensitivity, particularly at low luminances; M cells are up to 10 times more sensitive to light than P cells ( 224). While some have proposed that this reflected an inherent linkage between rod vision and the M cell pathways, more recent studies have shown that scotopic visual acuity at all eccentricities is better than could be supported by a mosaic of pure M cells ( 225). Hence this simplified viewpoint can not be substantiated. There seems to be a marked similarity between morphologic and functional characteristics of the M type ganglion cell in the primate and the Y type retinal ganglion cell in the cat. For example, in pri­mates, the temporal frequency response of M ret­inal ganglion cells is modified by increasing con­trast, and increasing the contrast modifies the tem­poral frequency response of Y retinal ganglion cells in the cat ( 226). In contradistinction, P retinal gan­glion cells in monkey and human gave a sustained response to increasing contrast. However, there are some who feel that P ganglion cells are not that similar to the cat X cells and instead should be regarded as a primate specialization not found in the retinae of lower mammals ( 226). It has also been found that, in monkey, there is a distinct pat­tern of electrotonic cellular coupling between M cells to each other and to amacrine cells; P ganglion cells however show no evidence of such coupling ( 227). Such M ganglion cell coupling may serve as a mechanism for increasing the luminance contrast sensitivity as well as variably maintaining large re­ceptive fields. While there may be relatively poor contrast sensitivity with P ganglion cells, this is in part compensated by the great number of P gan­glion cells in the retina. Recent studies have also validated Livingstone's proposal that color and motion may be processed in partially separate pathways with the P cell system mediating the former and the M cells system the latter ( 228,229). Livingstone and Hubel ( 222) proposed and more recently Merrigan and Maunsell ( 221) have shown that stereopsis, motion, and figure/ ground dis­crimination were probably M cell functions, whereas visual acuity and color vision were pri­marily P pathway mediated. While this might be an oversimplification of a highly complex set of systems, it at least sets the stage for understanding the segregation of two or more systems at both subcortical and cortical levels. However, at the subcortical level there are a number of other parallel pathways. Animal exper­iments in a variety of vertebrates suggest that there are retinal ganglion cell projections to at least five and probably eight or more primary visual nuclei. Retinofugal fibers have been clearly established to project to 1. the superior colliculus ( optic tectum); 2. the pretectum; 3. the lateral geniculate nuclei ( in most animals there is a separate projection to either the dorsal lateral geniculate nucleus or the ventral lateral ge­niculate nucleus); 4. the accessory optic system; and 5. the suprachiasmatic nucleus of the hypothal­amus ( 230). Projections to other nuclei of the hypothalamus have also been described in humans as well as in animals ( 231). Similarly, projections to the human accessory optic system have also been recently demonstrated ( 232). More recent studies have yielded insights into the development of these retinofugal pathways to the primary visual nuclei. It might be said that there is a paradigm of dendritic remodeling as a general mechanism for mammalian retinal gan­glion cell development by which extensive regres­sion of the dendritic tree occurs after birth ( 233); axonal elongation is followed by branching of col­laterals with final or elaboration of collaterals based upon the " correctness" of their connection ( 234); and electrophysiological restructuring of retinofugal projections can occur due to variations / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 ANTERIOR VISUAL PATHWAYS 243 of sodium- mediated action potentials ( 235). While these studies were done on rats, hamsters, and cats respectively, it was demonstrated in monkey that there is an order in which the retinal ganglion cells develop beginning with axons of small caliber and followed successively by medium size and fi­nally large retinal ganglion cells with large fibers ( 219). This is particularly interesting, given that the smallest class of cells contribute pathways to the hypothalamus and midbrain, which might be con­sidered more " primitive." Further evidence in the cat reveals development by fiber type ( chrono-topy), which might contribute to the morphologic differentiation of these different retinal ganglion cell types ( 236- 239). Both retinotopic and chrono-topic arrangements can be discerned not only at the level of each primary visual nucleus, but in the decussation patterns of fibers at the optic chiasm. While some feel that retinotopy ( preservation of the visual map) is maintained all along the fiber pathway until it hits its target visual nucleus, more recent studies have shown that HRP- labeled axon profiles emanating from a cluster of cat retinal gan­glion cells became quite scattered in the distal optic nerve ( 240). These studies have been repeated in monkey as well ( 241). Indeed, retinotopy as mea­sured by the number of fascicles is actually lost as the fibers progress from the retrobulbar optic nerve to the optic nerve near the optic chiasm ( 242). Moreover, fibers deriving from the temporal retina develop at a different time than those that derive from the nasal retina ( 243). Hence retino­topy or fiber arrangement must be maintained or restructured by more than simple mechanical and physical relations. How then do fibers know whether to decussate at the chiasm or not? One possibility is that each of the two paths of each retina ( retinal ganglion cells from the nasal half of the retina of each eye versus the temporal half) carry a different molecular trophic factor, which provides a lateralization signal at the level of the optic chiasm ( 244,245). Setting aside the problem of whether to decus­sate at the optic chiasm, the retinofugal axons de­riving from retinal ganglion cells must reorganize themselves and reestablish retinotopy at the level of the optic tract. This appears to occur suddenly at the optic chiasm/ tract junction and may also reflect some type of trophic molecular message ( 246). Substrates that might mediate such signals have been suggested in studies that attempt to regulate growth cone behavior of retinal ganglion cells grown in vitro ( 247). For example, extracts ob­tained from a suspension of cells from the anterior superior colliculus promoted the growth of retinal ganglion cells that derive from the temporal but not nasal retina ( 247). Hence it seems likely that at least each half of the retina contributes retinal gan­glion cells whose axons respond to a different mo­lecular gradient in a trophic fashion. Each gradient could establish an axis and indeed this has been demonstrated to exist in the optic chiasm of the ferret ( 248). Retinal ganglion cell axons, which ap­pear dispersed in the most posterior portions of the optic nerve, segregate sharply as they reach the ferret optic chiasm, such that fibers that derive from the dorsal retina immediately migrate caudo-medially, and those from the ventral retina rostro-laterally ( 248). Within the optic tracts, the largest fibers tend to run on the surface and the smallest at the core; this mirrors the arrangement of the optic nerves as well ( 249). This fiber arrangement de­scribed in the monkey probably explains why le­sions of the optic tract in humans produce such incongruous visual field defects. This arrangement of smaller axons centrally and larger axons more superficially may also reflect the chronotopic ar­rangement since small caliber retinal ganglion cells send off their axons first, while larger axons un­dergo axogenesis later ( 219). Another element that may help control chronotopicy may be the chang­ing glial environment along the optic pathways and particularly the optic tracts ( 250). The best understood primary visual nucleus re­mains the lateral geniculate nucleus. In the cat, retinotopically ordered retinofugal fibers project to the ventral and dorsal lateral geniculate nuclei; but only the dorsal lateral geniculate nuclei pro­ject to the visual cortex. Ventral lateral geniculate nucleus cells project to the pretectum and superior colliculus and probably other brainstem nuclei as well. Recent work has furthered this understand­ing of lateral geniculate nucleus function with em­phasis that it is more than a relay of monocular retinal information. In the dorsal lateral geniculate nucleus it has been shown that each neuron may be contacted by over 10 retinofugal axons and that each retinofugal fiber probably contacts more than 20 neurons in the lateral geniculate nucleus ( 251). Both M- and P- type retinal ganglion cells project to the dorsal lateral geniculate nucleus ( layers 1 and 2 for M cells and layers 3 and 5 for P cells). The third and smallest retinal ganglion cell type ( which we might regard as gamma or W- like) also projects to the dorsal lateral geniculate nucleus in primates ( 252). Recent updates on the circuitry of the lateral geniculate nucleus have confirmed and advanced the point first made by Guillery in 1971 in both cat and monkey that there is minimal binocular over­lap between contralateral and ipsilateral projec- / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 244 A. A. SADUN AND J. DAO tions to the lateral geniculate nucleus and that cells in each lamina fire only in response to incoming retinal signals ( 253,254). In contradistinction, the ventral lateral genicu­late nucleus, as described in the tree shrew, re­ceives input from the retina, cortex, pretectum, and superior colliculus ( 255). Two subdivisions of the ventral lateral geniculate nucleus probably me­diate visuosensory and visuomotor functions sep­arately; a third ventral lateral geniculate nucleus region projects to the suprachiasmatic nucleus of the hypothalamus, an important area for light en-trainment and synchronization of the circadian rhythm ( 256). In this sense, it can be seen that the lateral geniculate nucleus, if regarded as a cortical relay in rodents and cats, must also be regarded as a relay to brainstem visual nuclei. Despite the emphasis on the segregation of the M and P pathways, selective damage to one or the other pathway by ablation of the corresponding layers in the dorsal lateral geniculate nucleus in monkeys did not significantly affect the accuracy of ocular saccades to stationary or moving targets ( 257). Similarly, it was shown that either the M or P pathway alone was capable of relaying sufficient information about target motion to generate func­tionally effective smooth pursuit eye movements. In most mammals retinotopically ordered fibers project to the superficial gray layer of the superior colliculus with most of the contribution coming from the contralateral retina ( 258). In the monkey, unlike the rodent, most of the cells of the superior colliculus receive binocular information while es­tablishing a retinotopic map of the contralateral vi­sual field ( 259- 261). Specificity of connections and the role of chronotopy and position based hues, as well as neuronal activity and chemical gradients, have been explored through a variety of studies in rodents and other mammals ( 262- 267). In many ways, the patterning of retinotectal connections in rodents has been the model upon which specificity studies have been performed to determine how multiple cues work together to guide the sculpting of retinotopy. The striking neurochemical architec­ture of the superior colliculus as well as its absence of sulci has made it ideal for such study ( 266). While the superficial layer of the superior collic­ulus appears to be visuosensory in function, the deeper layer is probably multisensory and cer­tainly has some premotor functions. Between the two layers is an interlaminar zone with connec­tions that may be different in different mammalian species ( 268- 270). In this zone as well as in the deeper layers of the superior colliculus, neuronal activity has been observed in the initiation of a saccade as compared to that responsible for spec­ifying the horizontal and vertical amplitude of each saccade ( 271). The accessory optic system in the mesencepha­lon is comprised of thin fasiculi and associated nu­clei arranged differently in different vertebrates. The three or more clusters of cells that constitute the accessory optic system probably receive large-diameter retinofugal axons that are crucial for co­ordinating head and eye movements in order to maintain a stable gaze ( 230). Recent studies have established the nature of the GABAnergic circuits in the accessory optic system and the receptive field properties in mammals, including monkey ( 272,273). The histologic appearance of the three accessory optic system nuclei in humans is not sig­nificantly different from those of monkey ( 232). The retinofugal input to the pretectum has been well known in man because of its clinical signifi­cance for the pupillary reflex. It would be better to describe the pretectum as a complex of cell clusters rather than a single nucleus, all of which mediate the pupillary light reflex by relaying light informa­tion to the Edinger- Westphal nucleus. One of the pretectal nuclear subdivisions is the nucleus of the optic tract, which receives bilateral input and sub­serves optokinetic nystagmus ( 274,275). There is also evidence that the nucleus of the optic tract also helps along with the accessory optic system for gaze stabilization and control of visual-vestibular interaction ( 276). While the predomi­nance of retinal ganglion cell input to the nucleus of the optic tract is from small cells ( perhaps W- like cells), some medium and some large caliber axons also terminate in the nucleus of the optic tract ( 277). Direct retinal input has been demonstrated to the suprachiasmatic, the supraoptic, and the paraventricular nuclei of the hypothalamus in man ( 231). The suprachiasmatic nucleus in particular ( but probably all three of these hypothalamic nu­clei) helps subserve the light entrainment and syn­chronization of the circadian rhythm. In some an­imals, further visual input can be traced to the su­prachiasmatic nucleus from subdivisions of the lateral geniculate nucleus ( 278- 280). Most of the neurons of the suprachiasmatic nucleus are GABA producing, and there is good evidence that GABA should be considered the principal neurotransmit­ter of the circadian system ( 281). The suprachiasmatic nucleus and other hypo­thalamic nuclei can be considered as " biologic clocks" that use visual input as entrainment diur­nal rhythm. The output of the suprachiasmatic nu­cleus may contribute to the " biological calendar," / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 ANTERIOR VISUAL PATHWAYS 245 which expresses seasonal cycles and even mile­stones of development. For example, the volume of a human suprachiasmatic nucleus, and the number of vasopressin- immunoreactive neurons contained within it, fluctuate rhythmically over the year with values that begin twice as high in the autumn as in the summer ( 282). Indirect inputs between the suprachiasmatic nucleus and the pi­neal may help regulate the timing of developmen­tal milestones such as the onset of menarche. The pulvinar complex is found in the lateral as­pect of the thalamus, and lies above and somewhat medial to the lateral geniculate nucleus; it too has been shown to receive retinofugal projections in cat, monkey, and humans, respectively ( 283- 286). The function of the pulvinar remains poorly un­derstood but probably relates indirectly to the con­trol of eye movements by identifying the most sa­lient visual objects ( 287,288). The concept of parallel pathways has, especially in the last few years, proven to be a powerful par­adigm for understanding the organization of the visual system and, by consequence several inter­esting clinical features, especially in regard to methods of visual assessment for patients with vi­sual dysfunction. While it is somewhat simplified, it is useful to think of the M and P cell pathways as anatomically and functionally distinct and as chan­nels for different psychophysical features. The third or small cell pathways can be considered as a primitive system that projects to various brainstem nuclei that mediate unconscious visual informa­tion useful to the control of eye movements and the light entrainment of circadian rhythms. Certain diseases may selectively impair one pathway pver the other. For example, it has been demonstrated in Alzheimer's disease there is a predilection for involvement of the M pathway ( 289). Similarly, Quigley and his colleagues ( 290) have described greater losses of fibers in the M pathway in chronic glaucoma. This has been ana­tomically corroborated by electrophysiologic stud­ies ( 291,292) and histopathologic studies of the lat­eral geniculate nucleus ( 293,294). On the other side of the coin are diseases like optic neuritis, which may have predominant ef­fects on the P cell pathway. For example, optic neuritis patients who have recovered to normal vi­sual acuity and full fields may still have impair­ments of foveal critical flicker frequency, a psycho­physical test of visual temporal resolution ( 295). This may be because temporal resolution is higher in the M than in the P cell pathway. Similarly, Regan has shown that a patient with multiple scle­rosis and optic neuritis seemed to have impair­ments of recognition of motion- defined shapes, which are rendered visible by a shift in velocity ( 296). Once again, this suggests the selective im­pairment of P pathway retinal ganglion cell axons. However, using the simplified scheme of M and P cell circuitry to explain every clinical syndrome may be stretching things. For example, several au­thors have suggested that developmental dyslexia is caused by selective lesions to the M cell pathway and consequently selective injury to the magnocel-lular layers ( I and II) of the lateral geniculate nu­cleus ( 297- 300). Not withstanding their histopath­ologic and psychophysical studies, the clinical im­plications remain highly speculative, and there is the danger that special forms of " treatment" will be offered for profit in exploitation of these theo­ries and of the patients with dyslexia ( 301). The medical community must remain wary of inappro­priate application of preliminary scientific studies for justification of unproven treatment strategies. Blindsight is the ability of some apparently blind patients to react to visual stimuli even when they are totally unaware of conscious vision due to damage along the geniculate/ visual cortex path­way. It has been traditionally hypothesized that such abilities might be due to an alternate subcor­tical pathway ( i. e., superior colliculus). However, reexamination of this issue suggests that, at least in some patients, an isolated tiny island of residual vision not disclosed by conventional visual field examination, is responsible for providing sufficient visual information for accurate responses but in­sufficient for conscious awareness of vision ( 302). Notwithstanding these " reality checks," there can be no question that the existence of parallel path­ways provides important clinical insights, namely, that there is much more to vision than visual acuity and many more inputs to the brain than those sim­ply relayed through the P- cell lateral geniculate nucleus. It is quite likely that in the next decade advances in diagnosis and management of dis­eases will be made with a better understanding of these multiple and sometimes separate channels of visual information. REFERENCES 126. Beck RW, Cleary PA, Anderson MM Jr, et al. A random­ized, controlled trial of corticosteroids in the treatment of acute optic neuritis. N Engl J Med 1992; 326: 581- 8. 127. Cleary PA, Beck RW, Anderson MM Jr, et al. Design, methods, and conduct of the Optic Neuritis Treatment Trial. Controlled Clin Trials 1993; 14: 123- 42. 128. Beck RW. The Optic Neuritis Treatment Trial: implications for clinical practice. Arch Ophthalmol 1992; 110: 331- 2. 129. Beck RW, Cleary PA. Optic Neuritis Study Group. Optic / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 246 A. A. SADUN AND J. DAO Neuritis Treatment Trial: one- year follow- up results. Arch Ophthalmol 1993; 111: 73- 5. 130. Keltner JL, Johnson CA, Spurr JO, Beck RW. Optic Neu­ritis Study Group Baseline visual filed profile of optic neu­ritis: experience of the Optic Neuritis Treatment Trial. Arch Ophthalmol 1993; 111: 231- 4. 131. Keltner JL, Johnson CA, Beck RW, et al. Optic Neuritis Study Group: quality control functions of the visual field reading center ( VFRC) for the Optic Neuritis Treatment Trial ( ONTT). Controlled Clin Trials 1993; 14: 143- 59. 132. Pollock SC. Visual fields in optic neuritis. Reply to Keltner JL, Johnson CA, Spurr JO, Beck RW. Optic Neuritis Study Group baseline visual field profile of optic neuritis: expe­rience of the Optic Neuritis Treatment Trial. Arch Ophthal­mol 1993; 111: 1315- 6. 133. Keltner JL, Johnson CA, Spurr JO, Beck RW. Optic Neu­ritis Study Group baseline visual field profile of optic neu­ritis: experience of the Optic Neuritis Treatment Trial. Arch Ophthalmol 1993; 111: 1316- 7. 134. Beck RW, Kupersmith MJ, Cleary PA, Katz B. The Optic Neuritis Study Group. Fellow eye abnormalities in acute unilateral optic neuritis: experience of the Optic Neuritis Treatment Trial. Ophthalmology 1993; 100: 691- 8. 135. Optic Neuritis Study Group. The clinical profile of optic neurits: experience of the Optic Neuritis Treatment Trial. Arch Ophthalmol 1991; 109: 1673- 8. 136. Beck RW, Cleary PA, Optic Neurits Study Group. Recov­ery from severe visual loss in optic neuritis. Arch Ophthal­mol 1993; 111: 300. 137. Trobe JD. High- dose corticosteroid regimen retards devel­opment of multiple sclerosis in Optic Neuritis Treatment Trial. Arch Ophthalmol 1994; 112: 35- 6. 138. Lessell S. Corticosteroid treatment of acute optic neuritis. N Engl J Med 1992; 326: 634- 5. 139. Beck RW, Cleary PA, Trobe JD, et al. The effect of corti­costeroids for acute optic neuritis on the subsequent de­velopment of multiple sclerosis. N Engl J Med 1993; 329: 1764- 9. 140. Beck RW, Arrington J, Murtagh FR, et al. Brain magnetic resonance imaging in acute optic neuritis: experience of the Optic Neuritis Study Group. Arch Neurol 1993; 50: 841- 6. 141. Savino PJ. Optic Neuritis Treatment Trial: an editorial. J Neuro- ophthalmol 1994; 14: 55- 7. 142. Chrousos GA, Kattah JC, Beck RW, et al. Side effects of glucocorticoid treatment: experience of the Optic Neuritis Treatment Trial. JAMA 1993; 269: 2110- 2. 143. Bohuslavizki KH, Hinck- Kneip C, Kneip A, et al. Reduc­tion of MS- related scotoma by a new class of potassium channel blockers from Ruta graveolens. Neuro- ophthalmol 1993; 13: 191- 8. 144. Guy J, McGorray S, Fitzsimmons J, et al. Disruption of the blood- brain barrier in experimental optic neuritis: immu-nocytochemical co- localization of H202 and extravasated serum albumin. Invest Ophthalmol Vis Sci 1994; 35: 1114- 23. 145. Celesia GG, Brigell M, Gunnink R, Dang H. Spatial fre­quency evoked visuograms in multiple sclerosis. Neurology 1992; 42: 1067- 70. 146. Ashworth B, Aspinall PA. A longitudinal study of visual function in multiple sclerosis with a note on the Cam­bridge grating test. Neuro- ophthalmol 1993; 13: 135- 42. 147. Grochowicki M, Vighetto A, Berquet S, Pissavin C. Evol­utive patterns of the visual field in optic neuritis. Neuro-ophthalmol 1993; 13: 269- 73. 148. Miyazaki S, Fujiwara R, Shimo- oku M, Aoyama S. Visual prognosis in idiopathic optic neuritis patients with se­verely disturbed visual acuity. Neuro- ophthalmol Jpn 1993; 10: 15- 9. 149. Yamada K, Kiyosawa M, Yagi T, Tamai M. A case of child bilateral optic neuritis with magnetic resonance imaging findings. Neuro- ophthalmol Jpn 1992; 9: 90- 3. 150. Hedges TR III. Bilateral visual loss in a child with disc swelling. Comments by Sedwick LA, Slavin M. Surv Oph­thalmol 1992; 36: 424- 8. 151. Lossos A, Eliashiv S, Ben- Chetrit E, Reches A. Optic neu­ritis associated with familial Mediterranean fever. J Clin Neuro- ophthalmol 1993; 13: 141- 3. 152. Hamed LM, Silbiger J, Guy J, et al. Parainfectious optic neuritis and encephalomyelitis: a report of two cases with thalamic involvement. / Clin Neuro- ophthalmol 1993; 13: 18- 23. 153. DeBroff BM, Donahue SP. Bilateral optic neuropathy as the initial manifestation of systemic sarcoidosis. Am J Oph­thalmol 1993; 116: 108- 11. 154. Perlman JI, Forman S, Gonzalez ER. Retrobulbar ischemic optic neuropathy associated with sickle cell disease. / Neuro- ophthalmol 1994; 14: 45- 8. 155. Yohai RA, Bullock JD, Margolis JH. Unilateral optic disk edema and a contralateral temporal fossa mass. Am J Oph­thalmol 1993; 115: 261- 2. 156. Belden CJ, Hamed LM, Mancuso AA. Bilateral isolated retrobulbar optic neuropathy in limited Wegener's granu­lomatosis. J Clin Neuro- ophthalmol 1993; 13: 119- 23. 157. Kao LY, Chuang HC, Liang YS. Visual loss as the initial presentation of nasopharyngeal carcinoma. / Clin Neuro-ophthalmol 1993; 13: 24- 6. 158. Borruat FX, Schatz NJ, Glaser JS, et al. Visual recovery from radiation- induced optic neuropathy: the role of hy­perbaric oxygen therapy. / Clin Neuro- ophthalmol 1993; 13: 98- 101. 159. Moser FG, Rosenblatt M, De La Cruz F, et al. Embolization of the ophthalmic artery for control of epistaxis: report of two cases. Head Neck 1992; 14: 308- 11. 160. Nawa Y, Jaques JD, Miller NR, et al. Bilateral posterior optic neuropathy after bilateral radical neck dissection and hypotension. Graefes Arch Clin Exp Ophthalmol 1992; 230: 301- 8. 161. Brown RH, Schauble JF, Miller NR. Anemia and hypoten­sion as contributors to perioperative loss of vision. Anes­thesiology 1994; 80: 222- 6. 162. Sedwick LA. Getting to the heart of visual loss: when car­diac medication may be dangerous to the optic nerves. Comments by Hedges TR III, Newman NJ. Surv Ophthal­mol 1992; 366- 72. 163. Kumar A, Sandramouli S, Verma L, et al. Ocular etham-butol toxicity: is it reversible? J Clin Neuro- ophthalmol 1993; 13: 15- 7. 164. Weleber RG, Miyake Y. Familial optic atrophy with nega­tive electroretinograms. Arch Ophthalmol 1992; 110: 640- 5. 165. Kuroda N, Isobe M, Ishiwata Y, et al. A giant optic nerve glioma in an infant: case report. Neuro- ophthalmol Jpn 1991; 8: 385- 90. 166. Dutton JJ. Gliomas of the anterior visual pathway. Surv Ophthalmol 1994; 38: 427- 52. 167. Listernick R, Charrow J, Greenwald M. Emergence of op­tic pathway gliomas in children with neurofibromatosis type 1 after normal neuroimaging results. J Pediatr 1992; 121: 584- 7. 168. Cunliffe IA, Moffat DA, Hardy DG, Moore AT. Bilateral optic nerve sheath meningiomas in a patient with neuro­fibromatosis type 2. Br J Ophthalmol 1992; 76: 310- 2. 169. Dutton JJ. Optic nerve sheath meningiomas. Surv Ophthal­mol 1992; 37: 167- 83. 170. Lindblom B, Turwit CL, Hoyt WF. Optic nerve sheath meningioma: definition of intraorbital, intracanalicular, and intracranial components with magnetic resonance im­aging. Ophthalmology 1992; 99: 560- 6. 171. Eng TY, Albright NW, Kuwahara G, et al. Precision radi­ation therapy for optic nerve sheath meningiomas. Int J Radiat Oncol Biol Phy 1992; 22: 1093- 8. 172. Kalenak JW, Kosmorsky GS, Hassenbusch SJ. Compres­sion of the intracranial optic nerve mimicking unilateral normal- pressure glaucoma. / Clin Neuro- ophthalmol 1992; 12: 230- 5. 173. Smith JL. Compression of the intracranial optic nerve J Neuro- Ophthalmol, Vol. 14, No. 4, 1994 ANTERIOR VISUAL PATHWAYS 247 mimicking unilateral normal- pressure glaucoma: editorial comment. / Clin Neuro- ophthalmol 1992; 12: 236- 7. 174. Knapp MEP, Flaharty PM, Sergott RC, et al. Gaze- induced amaurosis from central retinal artery compression. Oph­thalmology 1992; 99: 238- 40. 175. Mansour AM, Dinowitz K, Chaljub G, Guinto FC. Meta­static lesion of the optic nerve. / Clin Neuro- ophthalmol 1993; 13: 102- 4. 176. Malik A, Furlan AJ, Sweeney PJ, et al. Optic neuropathy: a rare paraneoplastic syndrome. / Clin Neuro- ophthalmol 1992; 12: 137- 41. 177. Horton JC, Garcia EG, Becker EK. Magnetic resonance im­aging of leukemic invasion of the optic nerve. Arch Oph­thalmol 1992; 110: 1207- 8. 178. Siatkowski RM, Lam BL, Schatz NJ, et al. Optic neuropa­thy in Hodgkin's disease. Am J Ophthalmol 1992; 114: 625- 9. 179. Moshfeghi DM, Finger PT, Cohen RB, et al. Visual recov­ery after radiation therapy of orbital lymphoma. Am ] Oph­thalmol 1992; 114: 645- 6. 180. Kodsi SR, Younge BR, Leavitt JA, et al. Intracranial plasma cell granuloma presenting as an optic neuropathy. Surv Ophthalmol 1993; 38: 70- 4. 181. Casteels I, De Loof E, Brock P, et al. Sudden blindness in a child: presenting symptom of a sphenoid sinus muco­cele. Br J Ophthalmol 1992; 76: 502- 4. 182. Abe H, Hasegawa S Takagi M, et al. Contrast sensitivity of the stationary and drifting vertical stripe patterns in pa­tients with optic nerve disorders. Am } Ophthalmologica 1993; 207: 100- 5. 183. Abe H, Hasegawa S, Takagi M, et al. Temporal modula­tion transfer function of vision by pattern visual evoked potentials in patients with optic neuritis. Ophthalmologica 1993; 207: 94- 9. 184. Kardon RH, Haupert CL, Thompson HS. The relationship between static perimetry and the relative afferent pupil­lary defect. Am J Ophthalmol 1993; 115: 351- 6. 185. Cohen DB, Glasgow BJ. Bilateral optic nerve cryptococco­sis in sudden blindness in patients with acquired immune deficiency syndrome. Ophthalmology 1993; 100: 1689- 94. 186. Kestelyn P, Taelman H, Bogaerts J, et al. Ophthalmic man­ifestations of infections with Cryptococcus neoformans in pa­tients with acquired immunodeficiency syndrome. Am J Ophthalmol 1993; 116: 721- 7. 187. Newman NJ, Lessell S. Bilateral optic neuropathies with remission in two HIV- positive men. / Clin Neuro- ophthalmol 1992; 12: 1- 5. 188. Tomsak RL, Lystad LD, Katirji MB, Brassel TC. Rapid re­sponse of syphilitic optic neuritis to posterior sub- Tenon's steroid injection. / Clin Neuro- ophthalmol 1992; 12: 6- 7. 189. Quiceno JI, Capparelli E, Sadun AA, et al. Visual dysfunc­tion without retinitis in patients with acquired immuno­deficiency syndrome. Am ] Ophthalmol 1992; 113: 8- 13. 190. Tenhula WN, Xu S, Madigan MC, et al. Morphometric comparisons of optic nerve axon loss in acquired immu­nodeficiency syndrome. Am ] Ophthalmol 1992; 113: 14- 20. 191. Levin LA, Albert DM, Johnson D. Mast cells in human optic nerve. Invest Ophthalmol Vis Sci 1993; 34: 3147- 53. 192. Budenz DL, Farber MG, Mirchandani HG, et al. Ocular and optic nerve hemorrhages in abused infants with in­tracranial injuries. Ophthalmology 1994; 101: 559- 65. 193. Kattah JC, Chrousos GC, Roberts J, et al. Metastatic pros­tate cancer to the optic canal. Ophthalmology 1993; 100: 1711- 5. 194. Joseph MP, Lessell S, Rizzo J, Momose KJ. Extracranial optic nerve decompression for traumatic optic neuropa­thy. Arch Ophthalmol 1990; 108: 1091^ 3. 195. Levin LA, Joseph MP, Rizzo JF III, Lessell S. Optic canal decompression in indirect optic nerve trauma. Ophthalmol­ogy 1994; 101: 566- 9. 196. Girard BC, Bouzas EA, Lama SG, Soudant J. Visual im­provement after transethmoid- sphenoid decompression in optic nerve injuries. / Clin Neuro- ophthalmol 1992; 12: 142- 8. 197. Seiff SR. High dose corticosteroids for treatment of vision loss due to indirect injury to the optic nerve. Ophthal Surg 1990; 21: 389- 95. 198. Tang RA, Kramer LA, Schiffman J, et al. Chiasmal trauma: clinical and imaging considerations. Surv Ophthalmol 1994; 38: 381-^. 199. Neetens A, Traumatic bi- temporal hemianopia. Neuro-ophthalmol 1992; 12: 375- 83. 200. Hwang JF, Yau CW, Huang JK, Tsai CY. Apoplectic op-tochiasmal syndrome due to intrinsic cavernous hemangi­oma case report. / Clin Neuro- ophthalmol 1993; 13: 232- 6. 201. Kazim M, Kennerdell JS, Maroon J, et al. Choristoma of the optic nerve and chiasm. Arch Ophthalmol 1992; 110: 236- 8. 202. Liu GT, Lessell S. Spontaneous visual improvement in chiasmal gliomas. Am ] Ophthalmol 1992; 114: 193- 201. 203. Shutter LA, Kline LB, Fisher WS. Visual loss and a supra­sellar mass complicated by pregnancy. Tang R, Bedoya EM, Harper R, Hupp SL: Comments. Surv Ophthalmol 1993; 38: 63- 9. 204. Slavin ML, Lam BL, Decker RE, et al. Chiasmal compres­sion from fat packing after transsphenoidal resection of intrasellar tumor in two patients. Am } Ophthalmol 1993; 115: 368- 71. 205. Robinson DB, Michaels RD. Empty sella resulting from the spontaneous resolution of a pituitary macroadenoma. Arch Intern Med 1992; 152: 1920- 3. 206. Martin- Boglind LM. Computer- assisted interpretation of resolution visual fields from patients with chiasmal and retrochiasmal lesions. Ophthalmologica 1993; 207: 148- 54. 207. Gutman I, Melamed S, Ashkenazi I, Blumenthal M. Optic nerve compression by carotid arteries in low- tension glau­coma. Graefe's Arch Clin Exp Ophthalmol 1993; 231: 711- 7. 208. Chaturvedi N, Hedley- Whyte ET, Dreyer EB. Lateral ge­niculate nucleus in glaucoma. Am J Ophthalmol 1993; 116: 182- 8. 209. Von Noorden GK, Crawford MJ. The lateral geniculate nucleus in human strabismic amblyopia. Invest Ophthalmol Vis Sci 1992; 33: 2729- 32. 210. Gottlob I, Charlier J, Reinecke RD. Visual acuities and sco­tomas after one week levodopa administration in human amblyopia. Invest Ophthalmol Vis Sci 1992; 33: 2722- 8. 211. Kupersmith MJ, Vargas M, Hoyt WF, Berenstein A. Optic tract atrophy with cerebral arteriovenous malformations: direct and transsynaptic degeneration. Neurology 1994; 44: 80- 3. 212. Lindner K, Hitzenberger P, Drlicek M, Grisold W. Disso­ciated unilateral convergence paralysis in a patient with thalamotectal haemorrhage. / Neurol Neurosurg Psychiatry 1992; 55: 731- 3. 213. Bishop GH. Fiber groups in the optic nerve. A ] Physiol 1933; 106: 460- 70. 214. Polyak S. The retina. Chicago: University of Chicago Press, 1941. 215. Enroth- Cugell C, Robson GJ. The contrast sensitivity of retinal ganglion cells of the cat. / Physiol 1966; 187: 517- 52. 216. Boycott BB, Wassle H. The morphological types of gan­glion cells of the domestic cat's retina. / Physiol 1974; 240: 397- 419. 217. Perry VH, Oehler R, Cowey A. Retinal ganglion cells that project to the dorsal geniculate nucleus in the macaque monkey. Neuroscience 1984; 12: 1101- 23. 218. Silveira LCL, Perry VH. The topography of magnocellular projecting ganglion cells ( M- ganglion cells) in the primate retina. Neuroscience 1991; 40: 217- 37. 219. Raparort DH, JT Fletcher, La Vail MM. Genesis of neurons in the retinal ganglion cell layer of the monkey. J Comp Neurol 1992; 322: 577- 88. 220. Lynch JJ 3rd, Silveira LCL, Perry VH, Merigan WH. Visual effects of damage to P ganglion cells in macaque. Vis Neu-rosci 1992; 8: 575- 83. 221. Merigan WH, Maunsell JHR. How parallel are the primate visual pathways? Ann Rev Neurosci 1993; 16: 389- 402. / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 248 A. A. SADUN AND J. DAO 222. Livingston MS, Hubel DH. Psychophysical evidence for separate channels for the perception of form, color move­ment, and depth. / Neurosci 1987; 7: 3416- 68. 223. Atkinson J. Early visual development: differential func­tioning of parvocellular and magnocellular pathways. Eye 1992; 6: 129- 35. 224. Purpura K, Kaplan E, Shapley RM. Background light and the contrast gain of primate P and M retinal ganglion cells. Proc Natl Acad Sci, USA 1988; 4534- 7. 225. Lennie P, Fairchild MD. Ganglion cell pathways for rod vision. Vis Res 1994; 30: 477- 82. 226. Bernadette EA, Kaplan E, Knight BW. Contrast gain con­trol in the primate retina: P cells are not x- like, some M cells are. Vis Neurosci 1992; 8: 483- 6. 227. Dacey DM, Brace S. A coupled network for parasol but not midget ganglion cells in the primate retina. Vis Neurosci 1992; 9: 279- 90. 228. Kool FL, DeValois KK. The role of color in the motion system. Vis Res 1992; 32: 657- 68. 229. Dobkins DR, Albright TD. What happens if it changes color when it moves?: psychophysical experiments on the nature of chromatic input to motion detectors. Vis Res 1993; 33: 1019- 36. 230. Rodieck RW. Visual pathways. Annu Rev Neurosci 1979; 2: 193- 225. 231. Sadun AA, Johnson BM, Schaecter J. Neuroanatomy of the human visual system: Part III. Three retinal projection to the hypothalamus. Neuro- ophthalmology 1986; 6: 371- 9. 232. Fredericks CA, Giolli RA, Blanks RH, Sadun AA. The hu­man accessory optic system. Brain Res 1988; 454: 116- 22. 233. Yamasaki EN, Ramoa AS. Dendritic remodeling of retinal ganglion cells during development of the rat. / Comp Neu­rol 1993; 329: 277- 89. 234. Bhide PG, Frost DO. Stages of growth of hamster retinof-ugal axons: implications for developing axonal pathways with multiple targets. / Neurosci 1991; 11: 485- 504. 235. Skaliora I, Scobey RP, Chalupa LM. Prenatal development of excitability in cat retinal ganglion cells: action potentials and sodium currents. / Neurosci 1993; 13: 313- 23. 236. Ault SJ, Leventhal AG. Postnatal development of different classes of cat retinal ganglion cells. ] Comp Neurol 1994; 339: 106- 11. 237. Ault SJ, Thompson KG, Zhou Y, Leventhal AG. Selective depletion of beta cells affects the development of alpha cells in the cat retina. Vis Neurosci 1993; 10: 237- 45. 238. Nelson R, Kolb H, Freed MA. Off- alpha and off- beta gan­glion cells in cat retina. I: Intracellular electrophysiology and HRP stains. / Comp Neurol 1993; 329: 68- 84. 239. Kolb H, Nelson R. Off- alpha and off- beta ganglion cells in cat retina: II. Neural circuitry as revealed by electron mi­croscopy of HRP stains. / Comp Neurol 1993; 329: 85- 110. 240. Horton JC, Greenwood MM, Hubel DH. Nonretinotopic arrangement of fibers in an optic nerve. Nature 1979; 282: 720- 2. 241. Naito J. Retinogeniculate projection fibers in the monkey optic nerve: a demonstration of the fiber pathways by ret­rograde axonal transport of WGA- HRP. / Comp Neurol 1989; 284: 176- 86. 242. Evans A, Jeffrey G. The fascicular organization of the cat optic nerve. Exp Brain Res 1992; 91: 79- 84. 243. Reese BE, Colello RJ. Neurogenesis in the retinal ganglion cell layer of the rat. Neuroscience 1992; 46: 419- 29. 244. Reese BE, Baker GE. Changes in fiber organization within the chiasmatic region of mammals. Vis Neurosci 1992; 91: 527-^ 3. 245. Baker GE, Reese BE. Chiasmatic course of temporal retinal axons in the developing ferret. / Comp Neurol 1993; 330: 95- 104. 246. Reese BE, Baker GE, Tassinari G, Marzi CA. Fiber re­organization in the mammalian optic chiasm: De novo es­tablishment of the dorso- ventral retinal axis. Soc Neurosci Abstr 1991; 16: 1132. 247. Baier H, Bonhoffer F. Axon guidance by gradients of a target- derived component. Science 1992; 255: 472- 5. 248. Reese BE, Baker GE. The re- establishment of the represen­tation of the dorsoventral retinal axis in the chiasmatic region of the ferret. Vis Neurosci 1993; 10: 957- 68. 249. Reese BE. Clinical implications of the fiber order in the optic pathway of primates. Neurol Res 1993; 15: 83- 6. 250. Colello RJ, Buillery RW. Observations on the early devel­opment of the optic nerve and tract of the mouse. / Comp Neurol 1992; 317: 357- 78. 251. Robson JA. Qualitative and quantitative analyses of the patterns of retinal input to neurons in the dorsal lateral geniculate nucleus of the cat. / Comp Neurol 1993; 334: 324- 36. 252. Lachica EA, Casagrande VA. The morphology of collicular and retinal axons ending on small relay ( w- like) cells of the primate lateral geniculate nucleus. Vis Neurosci 1993; 10: 403- 18. 253. Garraghty PE, Sur M. Competitive interactions influenc­ing the development of retinal axonal arbors in the lateral geniculate nucleus. Physiol Rev 1993; 73: 529- 45. 254. Wilson JR. Circuitry of the dorsal lateral geniculate nu­cleus in the cat and monkey. Acta Anat 1993; 147: 1- 13. 255. Conley M, Friederich EB. Functional organization of the ventral lateral geniculate complex of the tree shrew ( Tal-spaia belangeri): II. Connections with the cortex, the thala­mus, and brainstem. / Comp Neurol 1993; 328: 21- 42. 256. Abe H, Rusak B. Stimulation of the hamster ventral lateral geniculate nucleus induces fos- like immunoreactivity in suprachiasmatic nucleus cells. Neurosci Lett 1992; 148: 185- 9. 257. Page WK, King WM, Merigan W, Maunsell J. Magnocel­lular or parvocellular lesions in the lateral geniculate nu­cleus of monkeys cause minor deficits of smooth pursuit eye movements. Vis Res 1994; 34: 223- 29. 258. Simon DK, O'Leary DDM. Development of topographic order in the mammalian retinocollicular projection. / Neu­rosci 1992; 12: 1224- 32. 259. Mark RF, James AC, Sheng XM. Geometry of the repre­sentation of the visual field on the superior colliculus of the wallaby ( Macrgous engenii). I. Normal projection. / Comp Neurol 1993; 330: 303- 14. 260. Marotte LR. Location of retinal ganglion cells contributing to the early imprecision in the retinotopic order of the developing projection to the superior colliculus of the wallaby ( Macropus eugenii). ] Comp Neurol 1993; 331: 1- 13. 261. James AC, Mark RF, Sheng XM. Geometry of the projec­tion of the visual field onto the superior colliculus of the wallaby ( Macropus genii) II. Stability of the projection after prolonged rearing with rotational squint. / Comp Neurol 1993; 330: 315- 23. 262. Molotchnikoff S, Itaya SK. Functional development of the neonatal rat retinotectal pathway. Brain Res Dev 1993; 72: 300- 4. 263. Harvey AR, Robertson D. Time- course and extent of reti­nal ganglion cell death following ablation the superior col­liculus in neonatal rats. / Comp Neurol 1992; 325: 83- 94. 264. Xiong M, Finlay BL. Changes in synaptic density after developmental compression or expansion of retinal input to the superior colliculus. } Comp Neurol 1993; 330: 455- 63. 265. Kondo Y, Takada M, Honda Y, Mizuno N. Bilateral pro­jections of single retinal ganglion cells to the lateral genic­ulate nuclei and superior colliculi in the albino rat. Brain Res 1993; 2: 204- 15. 266. Ming RB, Graybrel AM. Pattern formation in the develop­ing superior colliculus: Ontogeny of the periodic architec­ture in the intermediate layers. / Comp Neurol 1994; 340: 311- 27. 267. Galli- Resta L, Ensini M, Fusco E, et al. Afferent spontane­ous electrical activity promotes the survival of target cells in the developing retinotectal system of the rat. / Neurosci 1993; 13: 243- 50. 268. Mooney RD, Huang X, Rhodes RW. Functional influence / Neuro- Ophthalmol, Vol. 14, No. 4, 1994 ANTERIOR VISUAL PATHWAYS 249 of interlaminar connections in the hamster's superior col-liculus. / Neurosci 1992; 12: 2417- 32. 269. Behan M, Appell PP. Intrinsic circuitry in the cat superior colliculus: projections from the superficial layers. / Comp Neurol 1992; 315: 230- 43. 270. Hall WC, Lee P. Interlaminar connections of the superior colliculus in the tree shrew. I. The superficial gray layer. / Comp Neurol 1992; 332: 213- 23. 271. Glimcher PW, Sparks DL. Effects of low- frequency stim­ulation of the superior colliculus on spontaneous and vi­sually guided saccades. / Neurophysiol 1993; 89: 953- 64. 272. Vanderwant JJ, Nunes Cardoza JJ, Vander Togt C. GABAergic neurons and circuits in the pretectal nuclei and the accessory optic system of mammals. Prog Brain Res 1992; 90: 282- 305. 273. Hoffman KP, Distler C, Ilg V. Callosal and superior tem­poral sulcus contributions to receptive field properties in the macaque monkey's nucleus of the optic tract and dor­sal terminal nucleus of the accessory optic tract. / Comp Neurol 1992; 321: 150- 62. 274. Distler C, Hoffman KP. Early development of the subcor­tical and cortical pathway involved in optokinetic nystag­mus: the cat as model for man? Behav Brain Res 1992; 49: 69- 75. 275. Schweigart G, Hoffman KP. Pretectal jerk neuron activity during saccadic eye movements and visual stimulations in the cat. Exp Brain Res 1992; 91: 273- 83. 276. Cohen B, Reisine H, Yokota JI, Raphan T. The nucleus of the optic tract. Its functions in gaze stabilization and con­trol of visual- vestibular interaction. Ann NY Acad Sci 1992; 656: 277- 98. 277. Kato I, Okada T, Watanabe E, et al. Retinal ganglion cells related to optokinetic nystagmus in the rat. Acta Otolaryn­gol Stockh 1992; 112: 421- 8. 278. Morin LP, Blanchard J, Moore RY. Intergeniculate leaflet and suprachiasmatic nucleus organization and connec­tions in the golden hamster. Vis Neurosci 1992; 8: 219- 30. 279. Shibata S, Moore RY. Neuropeptide Y and optic chiasm stimulation affect suprachiasmatic nucleus circadian func­tion in vitro. Brain Res 1993; 615: 95- 100. 280. Laemle LK, Fugara C, Bentley T. The geniculolypotha-lamic pathway in a congenitally anophthalmic mouse. Brain Res 1993; 618: 352- 7. 281. Moore RY, Speh JC. GABA is the principal neurotransmit­ter of the circadian system. Neurosci Lett 1993; 150: 112- 6. 282. Hoffman MA, Swaab DF. Seasonal change in the supra­chiasmatic nucleus of man. Neurosci Lett 1992; 139: 257- 60. 283. Winborne BM, McCart RJ, Henry GH. Projections from the lateral division of the lateral posterior- pulvinar com­plex to area 21a and the striate cortex in the cat. Brain Res 1993; 603: 333- 7. 284. Hardy SG, Lynch JC. The spatial distribution of pulvinar neurons that project to two subregions of the inferior pa­rietal lobule in the macaque. Cereb Cortex 1992; 2: 217- 30. 285. Diamond IT, Fitzpatrick D, Conley M. A projection from the paratrigeminal nucleus to the pulvinar nucleus in Galago. / Comp Neurol 1992; 316: 375- 82. 286. Sadun AA, Johnson BM, Smythe LEH. Neuroanatomy of the human visual system. Part II. Retinal projections to the superior colliculus and pulvinar. Neuro- ophthalmology 1986; 6: 363- 70. 287. Robinson DL. Functional contributions of the primate pulvinar. Prog Brain Res 1993; 95: 371- 80. 288. Robinson DL, Petersen SE. The pulvinar and visual sa­lience. Trends Neurosci 1992; 15: 127- 32. 289. Sadun AA, Bassi C. Optic nerve damage in Alzheimer's disease. Ophthalmology 1990; 94: 9- 17. 290. Quigley HA, Dunkelberger GR, Green WR. Chronic hu­man glaucoma causing selectively greater loss of large op­tic nerve fibers. Ophthalmology 1988; 95: 357- 63. 291. Howe JW, Mitchell KW. Electrophysiologically deter­mined contrast sensitivity in patients with ocular hyper­tension and chronic glaucoma. Doc Ophthalmol 1992; 80: 31- 41. 292. Bray LC, Mitchell KW, Howe JW. Prognostic significance of the pattern visual evoked potential in ocular hyperten­sion. Br } Ophthalmol 1991; 75: 79- 83. 293. Charturvedi N, Hedley- White ET, Dreyer EB. Lateral ge­niculate nucleus in glaucoma. Am J Ophthalmol 1993; 116: 182- 8. 294. Dandona L, Hendrickson A, Quigley HA. Selective effects of experimental glaucoma on axonal transport by retinal ganglion cells to the dorsal lateral geniculate nucleus. In­vest Ophthalmol Vis Sci 1991; 32: 1593- 9. 295. Jacobson DM, Olson KA. Impaired critical flicker fre­quency in recovered optic neurits. Ann Neurol 1991; 30: 213- 5. 296. Regan D, Kothe AC, Sharpe JA. Recognition of motion-defined shapes in patients with multiple sclerosis and op­tic neuritis. Brain 1991; 114: 1129- 55. 297. Livingstone MS, Rosen GD, Drislane FW, Galaburda AM. Physiological and anatomical evidence for a magnocellular defect in developmental dyslexia. Proc Natl Acad Sci USA 1991; 88: 7943- 7. 298. Galaburda A, Livingstone M. Evidence for a magnocellu­lar defect in developmental dyslexia. Ann NY Acad Sci 1993; 582: 70- 82. 299. Galaburda AM. Neuroanatomic basis of developmental dyslexia. Neurol Clin 1993; 11: 161- 73. 300. Chase C, Jenner AR. Magnocellular visual deficits affect temporal processing of dyslexics. Ann NY Acad Sci 1993; 682: 320- 9. 301. Sadun AA. Dyslexia of the New York Times: ( Misunder­standing of parallel visual processing. Arch Ophthalmol 1992; 110: 933- 4. 302. Fendrich R, Wessinger CM, Gazzaniga MS. Residual vi­sion in a scotoma: implications for blindsight. Science 1992; 258: 1489. 303. Dandona L, Hendrickson A, Quigley HA. Selective effects of experimental glaucoma on axonal transport by retinal ganglion cells to the dorsal lateral geniculate nucleus. In­vest. Ophthalmol Vis Sci 1991; 32: 1593- 9. / Neuro- Ophthalmol, Vol. 14, No. 4, 1994