NATURE AND BASIS OF FUSION 333 



his bifocal spectacles. But there seems no reason for the evolution of a 

 special central mechanism for inhibiting any enhancement of visual acuity 

 that might accrue from the integration of the two monocular images. 



That integration must, then, be of a sort which somehow makes 

 impossible any real interdigitation of two complete monocular sets of 

 image points in a therefore-twice-as-well-resolved binocular image. This 

 condition will be satisfied if the fusion image somehow partakes of the 

 nature of a gross mosaic. And that it does do so, at least where fusion of 

 patterns is concerned, is suggested by the phenomenon of 'retinal rivalry': 



Suppose we observe in a stereoscope (Fig. 117, p. 316) a card, the two 

 pictures on which are like those in Figure 122a. We might reasonably 

 suppose that the two sets of diagonal lines would be fused into a perfect 

 grid; but they are not — what we see is a mosaic, composed from the two 

 sets of lines, which constantly shifts but which, at some one instant, might 

 look like Figure 122b. At no time do we see a standing grid pattern, 

 either throughout the whole square or even in some small area thereof. 

 Instead, the two unlike patterns vie for a place in consciousness, and at 

 any one time parts of each pattern are wholly successful. 



The image in such situations is generally deemed the very apotheosis 

 of a non-fusion image. But there has long been a theory, favored by a 

 minority of psychologists, that the everyday binocular image partakes of 

 the same ever-changing mosaic character as the rivalry image. It only 

 fails to exhibit rivalry (and hence fails to reveal its mosaic character) 

 because the two images being dovetailed together are identical or (where 

 the object is tridimensional) only slightly unlike — never as greatly dif- 

 ferent as are the two patterns of Figure 122a. Intra- and interhemispheric 

 fusions are thus essentially the same, for both involve putting left- and 

 right-eyed fragments side by side in the total image (Fig. 121a, p. 321). 



This mosaic theory of fusion has not yet had an adequate experi- 

 mental test, but it holds considerable promise. However, though it 

 accounts beautifully for the equality of binocular and monocular acuity 

 and brightness, it is helpless to explain the binocular mixture of colors. 

 One can obtain rivalry between, say, red and green monocular areas in a 

 stereoscope. But under proper conditions the red and green fuse into 

 homogeneous orange, which is not of heightened brightness, and yet has 

 no appearance of being a mosaic of red and green. It would seem that 

 the single images resulting from the binocular fusion of complementaries, 

 or of other miscible colors, must of necessity represent the fusion of all 

 of the right-eye image with all of the left-eye one. 



