18 ROYAL SOCIETY OF CANADA 



28. Upon three plates of sooted glass circles of 20, 40 and 60 mm,, 

 respectively, were traced. These were placed vertically, in succession, 

 on a sliding carriage, between a fishtail burner and the eyes, both 

 burner and glass being perpendicular to the line of sight. The centre 

 of the circular disc was " fixed " with the eyes for, perhaps, a quarter 

 of a minute or more, then the eyes were closed and all light excluded 

 and a negative after-image was seen. By trial, that position of the 

 carriage was then found by which the vibrating image completely filled 

 the circular disc of the negative after-image, without, however, extend- 

 ing beyond the limits of the disc. 



29. The results of these observations gave, as the retinal diameter 

 of the image when no light enters the eye, about 0-47 mm. According 

 to Kolliker the horizontal diameter of the yellow spot (Macula lutea) 

 is 3-24 mm., and the vertical diameter of the same 0-81 mm. These 

 measurements show that the image 'is seen well within the boundaries 

 of the yellow spot and, even in darkness, does not extend far beyond 

 the limits of the Fovea. 



30. In both of the foregoing experiments an attempt was made 

 to estimate the amplitude of the vibrations, this is, however, an exceed- 

 ingly difficult thing to do. In the first experiments two faint pencil 

 marks were drawn on the paper on either side of the centre to aid in 

 estimation, but in the second experiments no such expedient could be 

 adopted. The average results of these estimations are not to be relied 

 upon as being anything but crude; at the same time they are probably 

 not far from the truth. 



31. I find that the amplitude when light is entering the eye is about 

 one-tenth of the diameter of the image (i.e., about 0-018 mm.) ; and 

 that the amplitude when light is not entering the eye (i.e., in the after- 

 image) is, perhaps,^ one-twentieth of the diameter of the image (or, 

 about 023 mm.). 



32. What either the diameter of the image or its amplitude may 

 be in absolute darkness and with no after-image, it is hard to say, but 

 I do not think that the image becomes any larger than in the after- 

 image, or that the amplitude alters appreciably. 



33. Attempts to measure the amplitude by means of a background 

 vibrating with a simple harmonic motion, proved quite fruitless, although 

 the method has not yet been abandoned and may yet be productive of 

 accurate results regarding both the amplitude and the nature of the 

 vibrations. 



^ I say, " perhaps," because this value may be very inaccurate ; the amplitude, 

 in the case of the' after-image, especially, being difficult to estimate. 



