IMAGES FALLING ON THE KODS AND CONES. 769 



according to the probabilities, are not perfected, e.g., the end of a line or a human face. In 



other cases the condition known as " contraction" of the field of vision tends to fill up the gap. 



This will be evident on looking at the nine adjoining letters, so that 



e disappears ; we no longer see the three letters on each side of it in *L n 



straight lines, but b, /, h, d are turned in towards e. The adjoining ci D C 



parts of the field of vision seem to extend over and around the blind 



spot, and thus help to compensate for the blind spot. 



II. Optic Fibres Inexcitable to Light. The layer of the (J (q\ 

 fibres of the optic nerve in the retina is not sensitive to light. 



This is proved by the fact that, in the fovea centralis, which 



is the area of most acute vision, there are no nerve-fibres. s J^ 



Further, Purkinje's figure proves that, as the arteries of the " 



retina lie behind the optic fibres, the latter cannot be concerned in the perception 



of the former. 



III. Rods and Cones. The outer segments of the rods and cones have rounded 

 outlines, and are packed close together ; but natural spaces must exist betweeu 

 them, corresponding to the spaces that must exist between groups of bodies with a 

 circular outline. These parts are insensible to light, so that a retinal image is com- 

 posed like a mosaic of round stones. The diameter of a cone in the yellow spot is 

 2 to 2 *5 /J. (M. Schultze). If two images of two small points, placed very near each 

 other, fall upon the retina, they will still be distinguished as distinct images, pro- 

 vided that both images fall upon two different cones. The two images on the 

 retina need only be 3-4-5*4 /x apart, in order that each may be seen separately, 

 for then the images still fall upon two adjoining cones. If the distance be dimin- 

 ished so very much that both images fall upon one cone, or one upon one cone and 

 the other upon the intermediate or cement substance, then only one image is per- 

 ceived. The images must be further apart in the peripheral portion of the retina 

 in order that they may be separately distinguished. 



As the rounded end-surfaces of the cones do not lie exactly under each other, but are so 

 arranged that one series of circles is adapted to the interstices of the following series, this 

 explains why fine dark lines lying near each other appear to have alternating twists upon them, 

 as the images of these must fall upon the cones, at one time to the right, at another to the 

 left. 



IV. The fovea centralis is the region of most acute vision, where only cones are 

 present, and where they are very numerous and closely packed (fig. 521). The 

 cones are less numerous in the peripheral areas of the retina, and consequently vision 

 is much less acute in these regions. We may therefore conclude that the cones are 

 more important for vision than the rods. When we wish to see an object distinctly, 

 we involuntarily turn our eyes so that the retinal image falls upon the fovea 

 centralis. In doing this, we are said to "fix " our eyes upon an object. The line 

 drawn from the fovea to the object is called the axis of vision (fig. 557, S r). It 

 forms an angle of only 3 '5-7 with the "optical axis" (0 A), which unites the 

 centres of the spherical surfaces of the refractive media of the eye. The point of 

 intersection, of course, lies in the nodal point (Kn) of the lens (p. 770). The term 

 " direct vision " is applied to vision when the direction of the axis of vision is in 

 line with the object, [i.e., when the image of the object falls directly on the fovea 

 centralis.] 



" Indirect vision " occurs when the rays of light from an object fall upon the 

 peripheral parts of the retina. Indirect vision is much less acute than the direct. 



To test the acuity of direct vision, draw two fine parallel lines close to each other, and 

 gradually remove them more and more from the eye, until both appear almost to unite and 

 form one line. The size of the retinal image may be ascertained by determining the distance 

 of the two lines from each other, and the distance of the lines from the eye, or, from the cor- 

 responding visual angle, which is generally from 60 to 90 seconds. 



Perimetry. In order to test indirect vision, we may use the perimeter of Aubert and 

 Forster. The eye is placed opposite a fixed point, from which a semicircle proceeds, so that 

 the eye lies in the centre of it. As the semicircle rotates round the fixed point, on rotating the, 



3c 



