404 Hastings — Optical Errors of the Human Eye. 



| K— » 1 )tya+-fc-- w 1 )fy r i I 



If this quantity be divided by the distance from n 2 to the 

 retina, we shall have its angular width. This is found to be 

 equal to 1', a conspicuous magnitude, since alternate bright and 

 dark lines of half this angular width can be distinguished 

 under favorable circumstances. 



The second part of the problem is not quite definite because 

 the radius of the pupil is not a constant; but if we assume 

 a diameter equal to 02 cm we shall be near the true value for 

 ordinarily bright illumination. This assumption gives 6'*8 for 

 the angular value of the radius of the diffusion circles ; in 

 other words, this is the angular width of the strip within 

 which the illumination of the retina by blue light falls from 

 its maximum to zero. As experiment shows that a diminution 

 of luminous intensity of less than one per cent is obvious at 

 a sharply defined border, there is no room for doubt that the 

 peculiarity of construction does possess a useful function in 

 vision, inasmuch as that with the established collimation errors 

 we are enabled to detect boundaries of colored fields with a 

 degree of precision which would be wanting in eyes without 

 such errors. 



This is a highly interesting reason for the persistence of a 

 systematic optical error in the human eyes which is otherwise 

 extremely difficult to account for, since no other possible error 

 admits of correction so easily. If we recognize the advantage 

 which the peculiar relation of the collimation errors in the two 

 eyes gives to one searching for colored fruits or colored animals, 

 we may be led to the conclusion that in an earlier state of 

 racial development the peculiarity would have been more 

 important than at present, and then, perhaps, to speculation as 

 to whether it may not now be regarded as largely vestigial. 

 This last idea would doubtless find support in the extraordinary 

 irregularity of the constant of collimation in the sixty or more 

 cases recorded. 



II. Geranium Phenomenon. 



This is a name given provisionally to a peculiar visual phe- 

 nomenon with which I have been familiar for an indefinite 

 time, although it does not seem to have ever been noted, or at 

 least recorded, by other observers. It is not unlikely that 

 relatively few persons are able to see it, but it is not a personal 

 peculiarity. Briefly described the phenomenon is this : When- 



