The absence of diplopia in squint may be explained quite simply by the fact that the habit of binocular fixation has not been learnt or has been forgotten; one can learn nothing that cannot be again forgotten. The normal fusion of the visual fields can only develop in consequence of binocular fixation, and diplopia is only possible when some kind of binocular fusion exists. If no binocular fusion exists, then all possibility of diplopia is excluded. And why should those who squint from their earliest childhood not see well with both eyes, but yet with each separately, just as is the case with animals with laterally placed eyes? For example, in Fig. 3 there is convergent squint of the left eye, the right eye fixes the point a, whose retinal image is cast at _a_' in the left eye; the direction outwards in which these images are projected is discovered by drawing a straight line from _a_ to _c_ (the optical centre of the eye); suffice it to say that point _a_ is seen by each eye in the direction in which it really stands. [Illustration: FIG. 3.] But although both eyes see at the same time, yet the close relation which in binocular fixation develops between the centres of the retinæ does not occur in squint; firstly, because the retinal area in the squinting eye which corresponds to the fixation point is too excentric, and secondly, because the angle of the squint often changes. In binocular fixation, the fixation point of one retina answers to the corresponding point of the other; in squint, on account of the varying size of the squint angle, if a like relation develops between the eyes, the fixation point of one retina must correspond to a larger area of the other. Possibly this explains a fact that is often to be observed. In those cases of squint where diplopia can easily be caused by covering one eye with a red glass and the other with a vertically deviating prism, the double images disappear on rotation of the prism round the axis of vision, as soon as the angle of the prism reaches an angle of about 45°. The occurrence of double images shows that there exists for the upper and lower parts of the retinæ a community of vision by no means coinciding with the identity principle. The disappearance of the diplopia can be explained by the fact that the variations of the squint angle take place chiefly in the horizontal direction. Therefore the area in the squinting eye that corresponds to the fovea centralis of the fixing eye must be more extensive in the horizontal than in the vertical direction. Alfred Graefe has designated this phenomenon as "regional exclusion." Whilst then a sort of community of vision exists for the upper and lower parts of the retinæ, the sensations of the retinal area lying in the horizontal plane of the macula lutea of the squinting eye must be suppressed. The physiological occurrence of a suppression of the retinal images, as far as we are able to investigate it, always refers