Hastings — Optical Errors of the Human Eye. 409 



IV. Binocular Color Relief. 



This is an optical illusion not noted in Helmholtz but 

 described and studied at length by W. Einthoven, who was 

 incited to the investigation by Donders. * According to this 

 writer, Professor Donders was the first one to record the fact 

 that red and blue objects in the same plane appear to most 

 observers, when looked at with both eyes, to lie in different 

 planes, especially when on a black background. To him, as 

 well as to Einthoven, the red appears nearer than the blue. 

 The explanation offered by Einthoven can be described by 

 reference to fig. 1 above. Imagine both axes of the vision 

 converged towards a red point which is accompanied by a blue 

 point just above it : the red images would fall on the points of 

 distinct vision, while the centers of the diffusion circles for 

 blue would fall on disparate points of the retinas nearer the 

 median plane, just where sharp images of a more distant red 

 point would fall ; hence the interpretation of the resulting 

 sensation is that the blue point is in fact more remote. He 

 finds evidence in favor of this view by the ingenious experi- 

 ment of cutting off, by means of movable diaphragms in front 

 of the eyes, first the inner halves of the pupils and then the 

 outer ; by this means he enhances or inverts the illusion. His 

 experiments are easily repeated and are highly interesting, f 



There are, however, serious objections to this explanation, 

 not only as accounting for the observations, but also from 

 theoretical considerations. Of about thirty individuals inves- 

 tigated Einthoven found that nearly one-third were not con- 

 fident at first that there was any sensation of relief, while 

 ultimately about half recognized the relief as did Donders, and 

 the other half inferred an inverted order. This general con- 

 clusion is fairly well supported by my own more limited 

 experiments. According to the theory, we are obliged to con- 

 clude that half of those investigated had the centers of their 

 pupils lying on the nasal side of the axes of vision, and half 

 on the temporal. It is, however, quite certain that a displace- 

 ment of the pupil towards the nasal side of the axis of vision 



* Stereoskopie durcli Farbendifferenz. Graefe's Arch., xxxi (3). 



f A convenient and interesting modification of this experiment may be 

 made by observing a flat surface divided into separated areas of saturated 

 colors through a binocular telescope of which the separation of the axes can 

 be varied at will. The familiar prismatic binoculars are best adapted to the 

 purpose, and any mosaic glass window in which the colors are vivid forms 

 a far better object than colored pigments. If the axial separation is a little 

 greater than that of the eyes, a striking relief will be observed in the sense 

 described above, while a smaller separation will invert the relief. 



