LETTER TO THE EDITORS A CHART DEMONSTRATING VARIXTiONS IN ACUITY WITH RETINAL POSITION Visual acuity

Visual acuity is d&ned as the reciprocal of the visual angle, in minutes. subtended by a just resolvable stimulus. This critical dimension can be the gap in a Landolt C, the spatiat period of a grating or the offset of a vernier. etc. Acuity declines progressively from the fovea out to the periphery of the retina. It is down to half its maximum value at an eccentricity of only 30 min according to Jones and Higgins (1947). who used Landolt C’s or 105 min according to Ludvigh (1941). who used Snellen letters. Weymouth (1958) plotted the reciprocal of visual acuity. namely minimum angle of resolution, and found that several visual thresholds increased linearly with increasing retinal eccentricity. at least up to 30’. This was true of the data of Ludvigh (1941) on acuity for Snellen letters, Wertheim (1893) for gratings. Bourdon (1902) for vernier offsets, and Basler (1906. 1908) for movement detection thresholds. In step with these variations in acuity, the physiological grain becomes progressively coarser towards the periphery of thr retina-receptor density, receptive field size and amount of visual cortex available. Hallett’s review (1963) and Wilson’s experiments (1970) showed that retinal summatory areas increase progressively with eccentricity. Hubel and Wiesei (1960) found that the mean diameter of receptive fields increases more or less lineariy with eccentricity in the monkey retina. The same was true for the number of min arc of visual fieid handled by each mm of visual cortex (Daniel and Whitteridge 1961). Sixty-seven microns of cortex. containing about live cells, corresponded to one threshold unit anywhere in the retina. It appears to take a constant number of retinal receptive fields, and a constant amount of cortex. to do an equaliy fine discrimination. whether in the fovea or the periphery. We measured recognition thresholds for letters (Paratipe Helvetica medium typeface) at retinal eccentricities lying between 4’ and 55’. Two subjects were used, both of whom wore correcting eyeglasses. They viewed binocularly a white tangent screen of luminance 1% log ft-L from a distance of 57 cm. Individual letters were moved inwards on the horizontal meridian from the periphery towards the fovea until they were correctly identified. Trials on which the subject said he had involuntarily gIanced towards the target letter were discarded. Results are shown in Fig. 1. Recognition thresholds increased linearly with eccentricity up to 30’. with a 0 10 20 30 40 50 60