iioo VISION. 



a lens on the sclerotic, the point of light being as fine as possible and at 

 some distance from the cornea (Fig. 393) ; by illuminating the eye laterally, 

 so that the light enters the pupil and produces an image on the nasal 

 side of the retina, while the light from this lateral image in its turn 

 throws shadows of the retinal vessels (Fig. 394) ; and, thirdly, by holding a 

 card against a brightly illuminated background, with a small hole in the 

 anterior focus of the eye. In each case the source of illumination should 

 be moved ; by the first and third methods the apparent vertical movement 

 of the figures will be in the same direction as the movement of the source 

 of illumination ; by the second method it will be in the opposite direction. 

 In calculating the position of the sensitive elements, the first method is 



the most satisfactory, and was used by Heinrich 

 y Miiller, to whom we owe the experiment. In 

 Fig. 393, a|3 may be calculated directly from AB, 

 the apparent movement on a screen at a known 

 distance ; if ij3 and ace be joined, their point of 

 intersection, v, will give the position of the 

 vessel. Miiller measured the distance, bv and av, 

 on frozen eyeballs, and took the mean of the 

 variations met with ; the distances, ccv and fiv, 

 could then be calculated. In his own eye the 

 distance thus ascertained varied from 17 to '32 

 mm., for the eye of three other observers from 

 Fig. 394. . 19 fco . 33 The distance b e t we en the vessels and 



the rod and cone layer in man was found to be between -2 and 3 mm. 



Konig and Zumf t l have made use of a modification of Purkinje's third 

 method to determine the depth of the sensitive elements for white and 

 monochromatic lights. They held a screen with two holes closely before 

 the eye, and obtained two shadow figures of the same vessel. These were 

 projected on a scale, and the depth of the sensitive layer calculated from 

 the distance of the projected figures on the scale and that between the 

 two holes. They give the following figures : — For white light the dis- 

 tance from the vessel to the sensitive layer was -4120 mm. ; for light of 

 670 X, 4402 mm. ; of 535 X, '4141 mm. ; of 434 X, -3643 mm. Satis- 

 factory results were only obtained with one observer. They seemed to 

 show that the greater the wave-length, the more deeply seated is the 

 sensitive layer. The difference between their extreme measurements, 

 namely, -0759 mm., is rather more than the thickness of the rod and cone 

 layer, and Konig supposes that red light may act on the retinal pigment. 



Leber and Koster 2 were unable to obtain two images with two holes, 

 as in Konig's method, and also failed to confirm his results by another 

 method. From Konig's results it should follow that Purkinje's images, 

 projected on a field half red and half blue, should show a difference of 

 parallax ; their relative rapidity should be as 22 : 19. They failed to 

 observe any difference, although they found by separate experiments that 

 a difference of 22 : 19 in rapidity was capable of being perceived. 



Evidence as to the seat of the sensitive elements of the retina has 

 also been derived from the degree of delicacy of local discrimination in 

 the retina. The visual angle necessary to enable two points to be 

 distinguished has been tested by observations on stars and on fine 

 parallel lines. The figures obtained by different observers vary from 50" 



1 Sitzungsb. d. k. Akad. d. Wisscnsch. , Berlin, 1894, S. 439. 



2 Arch./. Ophth., 1895, Bd xli. Abth. 1, S. 1. 



