SUMMARY1. The effects of picrotoxin and strychnine were tested on the receptive fields of direction sensitive cells, orientation sensitive cells, local edge detectors, uniformity detectors and large field units in the rabbit retina.2. Picrotoxin eliminated the direction specificity and size specificity of 'on-off' and 'on' directionally sensitive cells for both black and white objects. Picrotoxin also made 'on' directionally sensitive cells responsive to faster velocities.3. Picrotoxin eliminated the orientation specificity of orientation sensitive cells, and changed the bar-flank arrangement of the receptive field into a centre surround arrangement. Thus, the orientation specificity is due to inhibitory rather than excitatory mechanisms.4. Picrotoxin altered the speed sensitivity of large field units so that they responded to slow speeds as well as fast ones, like centre surround Y cells.5. Strychnine abolished the size specificity of local edge detectors and changed their speed specificity so that they responded to faster speeds.6. Picrotoxin changed a uniformity detector into a sustained on centre cell. 7. Strychnine did not affect the direction specificity of directionally sensitive cells, the orientation specificity of orientation sensitive cells, or the speed specificity of large field units. Picrotoxin did not affect the size specificity of local edge detectors.8. Picrotoxin and strychnine usually had opposing effects on the transient responses of these units to spots and annuli. In general picrotoxin prolonged and enhanced these responses at both on and off, and strychnine shortened them.9. The effect of these drugs for every type of ganglion cell with complex receptive field properties was to make the receptive field more simple. The orientation selective cells, large field cells, 'on' direction selective cells and uniformity detectors seem to be centre surround cells with special properties that are abolished by these drugs. The 'on-off' direction selective cells and local edge detectors still have on-off receptive fields, but in each case one of the drugs abolished the feature that was the basis for the cell's name.

The receptive fields of directionally sensitive ganglion cells in the rabbit retina were analyzed. Several types of experiment showed that each point within the receptive field of the cell is inhibited by a fairly wide area of points around it, lying on each side of the preferred-null axis as well as along the preferred-null axis in the preferred direction. The excitatory or responsive receptive field of these cells has an inhibitory surround: this inhibitory surround appears to be simply an extension of the inhibition that occurs within the center of the receptive field. Points toward the edge of the responsive receptive field are inhibited from an area around them which extends into the center of the receptive field and also into the inhibitory surround. Directionally sensitive retinal ganglion cells respond to moving spots better than to moving bars. This is particularly true for objects moved perpendicularly to the preferred-null axis. In some cells a spot moved perpendicularly to the preferred-null axis will give a substantial response, whereas a bar moved in the same direction will give no response at all. This phenomenon can be explained by the inhibitory area which surrounds each point within the receptive field; since this inhibitory area is asymmetrical, it is also responsible for the cell's directional sensitivity. When two bars oriented perpendicular to the preferred null axis are flashed, one after the other, the response to the second bar is nearly always reduced by the presentation of the first bar. This is true for many temporal and spatial sequences corresponding to movement in the preferred direction, as well as those corresponding to movement in the null direction. However, there are temporal and spatial sequences, corresponding to movement in the preferred direction, for which the response to the second bar is unaffected by the presentation of the first bar. The time delay for this does not vary from cell to cell--it is always approximately 20 ms for on-off directionally sensitive cells and approximately 180 ms for on directionally sensitive cells. The spatial separation does vary from cell to cell, between 0.13 degrees and 1.2 degrees in 11 on-off directionally sensitive cells. This spatial separation, which gives linear summation of the response to two bars flashed 20 ms apart in the preferred direction, is correlated with the speed of movement which gives the best response for a bar moved through the receptive field in the preferred direction.

The purpose of this study was to characterize the form of the pupil in normal human subjects. Using a modified slitlamp, photographs of pupils were taken in steady illumination and 10-20 sec after darkness. Transparencies were projected and digitized, and the pupil margin was represented as a circular Fourier series. Best-fit ellipses were also determined. The placement of the pupil relative to the limbus was determined in a number of subjects. Results from 23 subjects indicated that in both darkness and light, average pupil noncircularity was 0.0166. (A value of 0.0200 is easy to detect with the unaided eye from the photographs.) On average, the best-fit ellipse accounted for about half of the noncircularity (59.6% in darkness; 47.7% in light). Most of the contribution to shape was made by the first 4 or 5 harmonics. Shapes were usually stable within a session and could remain fairly stable for at least a year; however, shapes for different subjects were not very similar, especially in the light. (Average pairwise similarity: 0.106 in darkness; 0.034 in light; similarity can have values from -1 to 1.) For a given subject, shapes in light and dark were often fairly similar (average similarity 0.260), but there were systematic differences: in eyes where the ellipse contributed > 20% of noncircularity, ellipse major axes clustered around vertical in darkness, and horizontal in light, implying greater contraction near the vertical meridian. Even pupils with little elliptical contribution turned out to contract more near the vertical meridian. There was some tendency for left and right eyes of an individual to show mirror symmetry of shape. In the dark, pupils were located 0.27 +/- 0.09 mm nasal and 0.20 +/- 0.15 mm superior to the limbus center, and usually moved slightly further nasal or superior in the light. Noncircularity increased with age (0.0015/decade). It was concluded that pupils show individuality of shape, with significant regularities within and across subjects.

Directionally sensitive ganglion cells in rabbit retina lose their directional sensitivity when picrotoxin, an antagonist of the inhibitory neurotransmitter gamma-aminobutyric acid, is infused into the retinal blood supply. Strychnine, an antagonist of glycine, does not produce this effect. Other receptive field types are affected by strychnine but not picrotoxin. Inhibitory transmitters therefore have specific functions in information processing in the retina.

Video-based devices for measuring gaze direction are widespread. However, there is a built-in imprecision in such devices in the event that pupil diameter changes during the experiments. Data are presented to demonstrate this effect. The possibility of correcting eye-position records for the imprecision is discussed and preliminary examples of such correction are presented.