Hastings — Optical Errors of the Human Eye. 405 



ever I look at red geranium petals, in the brilliant light of out 

 of doors, projected against the more remote background of its 

 green leaves, the petals seem — in portions of the visual field 

 at any rate — to be bordered with an exquisitely fine line of 

 intense blackness, much more intense than that of black velvet 

 under the same illumination. My eyes cannot make these 

 black lines point of fixation for they are singularly elusive, but 

 they are too delicate to be seen if they did not fall very near 

 the axis of vision. Red petals of other flowers exhibit quite 

 the same phenomena, notably red nasturtiums, but only strik- 

 ingly when the green background is sufficiently luminous. 



The more important general conclusions from such observa- 

 tions seem to be embodied in the following list : 



(a) The phenomenon is monocular. 



(b) The dark lines are conspicuous only when the illumina- 

 tion is intense (equals, probably, when the pupillary aperature 

 is small). 



(c) The contrasting colors must be well separated as regards 

 refrangibility and of approximate equal brightness. 



(d) The difference of the distances of the two colored fields 

 from the eye must not be small compared to the distance of 

 the nearer one. 



By attention to these precepts I have been able to observe 

 the same phenomenon with a considerable range of pure spec- 

 tral colors as well as to prove that the order of the colors, as 

 measured from the eye, is not essential. The reason that I see 

 the black lines as described above, and no black border to a 

 green leaf projected upon a red petal, is to be ascribed, prob- 

 ably, to the fact that my slightly myopic eye can form a sharp 

 retinal image of the red when not more than one or two meters 

 distant, while the green leaf would be notably out of focus. 



A highly probable explanation of this phenomenon can, I 

 think, be found from an inspection of , fig. 1 above. Suppose 

 the parti-colored object be shifted so that the line of demarca- 

 tion falls on the line n 1 ir 1 extended ; then, in the right eye, r 

 and b 1 will have approached each other, but will still be dis- 

 tinct. Now imagine the blue portion of the object carried a 

 considerable distance farther from the eye on the line n 1 tt } 

 extended, then light from points in the blue object very near 

 its edge will send light to only the outer, or temporal, half of 

 the pupil, the inner half being shaded by the nearer red object. 

 It is, however, just this latter half of the pupil that transmits 

 the light which diffuses blue light on the temporal side of the 

 line 7r 2 b„ hence there is a region of the retina between r and 

 b 1 in the right eye, upon which, under the conditions considered, 

 no light falls on any color whatever. No doubt such an 



