The visual image is throughout determined by tactual experience. With each feature of tactual extension there is correlated a corresponding feature of visual extension. The visual image is throughout moulded by the touch. Now we are here concerned, not with visual imagery, but with visual perception. What we say is that owing to the frequent and intimate union of tactual with visual perceptions, the visual perception, when it exists without the tactual, will be moulded by previous tactual experiences, much as the visual image of an object in the dark is moulded by a present tactual experience. This is only a special application of the general principle which underlies the whole development of the spatial perception. In the development of the tactual perception of space passive or synthetic touch acquires a certain serial order and arrangement of parts from its connexion with active and analytic touch. In like manner, the visual perception of extension acquires a certain order and arrangement of its parts from its connexion with the tactual perception of extension. It would thus appear that though the eye had no independent means of apprehending those relations of surfaces and lines which presuppose the third dimension, it would none the less become capable of apprehending them in some degree through its intimate practical union with touch. The whole process is a case of complication.* It is difficult to say precisely how much of our perception of solid figure and distance is gained in this way, and how much is due to other factors. But there does not seem much room for doubt that the combination of visual and tactile experience plays a part of primary importance. What it does not account for is the great superiority of vision over touch in the apprehension of the distances of objects from each other and from the body. To explain this, we must take into account conditions peculiar to vision. Among these the most important is connected with the fact that we normally use two eyes to see, and not one only.

(b) As Dependent on the Use of Two Eyes. Whenever we look at an object with both eyes, we receive from it two impressions, one affecting each eye. But the result is a single presentation of the object. This is so because similar impressions fall on corresponding points of the two retinas. The two points of most distinct vision constituted by the foveae centrales, or central pits, correspond to each other in this way, so that lightimpressions falling on them give rise to the vision of a single object. Other points of the two retinas also correspond when they are symmetrically situated with reference to the central pit. In general, the left half of one eye corresponds to the left half of the other, and the right half to the right half. Thus a point in the left half of one eye will correspond to a point in the left half of the other when both have the same situation relatively to the centre of distinct vision. If the retina of one eye could be applied to the retina of the other, so as to superpose the nasal half of each on the temporal half of the other, their points of contact would be, roughly speaking, corresponding points. Single vision occurs when corresponding points are stimulated in a similar way. But it also occurs when the points thus stimulated do not exactly correspond, but when the deviation from correspondence, or disparateness, as it is called, is small. When this happens, a single object is seen, but it is seen as lying behind or before that area of the field of vision which is most distinctly presented, or, in other words, that part of the field of vision which we are directly looking at. When the disparateness between the points affected is relatively great, double vision may result.

If a finger is held between the eyes and an object, and if we then fix our eyes on the object so as to bring it into the centre of distinct vision, we see the finger doubled. The greater the distance between the finger and the object, the wider apart are the two images of the finger. If the right eye is closed, the left image disappears; if the left eye is closed, the right image disappears. If now, instead of fixing our gaze on the object, we fix it on the finger, the finger is seen as single, and the object as double. The greater the distance between finger and object the wider apart are the two images. When the right eye is closed, the right image disappears. When the left eye is closed, the left image disappears. This experiment succeeds with most people, but not with every one. There are some few who can hardly be brought to see things double at all. But even these, if they really use their two eyes in a normal way, will in all probability be able to see a double image of such a bright object as a lighted candle. The special conditions of the experiment are that the eye should be fixed on one object, and the attention fixed on another, either beyond it or in front of it. Apart from these special conditions, it would appear that double images are not ordinarily discerned by normal persons. When the eyes are moving in a free and natural way from object to object, and attention is concentrated only on what is seen in the area of distinct vision, double images are not discerned. It should be carefully noted that when vision is distinctly doubled, the distance of the two images from the object which is distinctly seen is very indeterminately apprehended. We may see it now at one distance and now at another, either arbitrarily, or in consequence of some casual suggestion. So far as the distance is determinately apprehended, our perception of it may be referred to other conditions than the disparate position of the two impressions on the retinas of the two eyes. It would seem that this disparateness either yields a definite perception of distance, or a double image, but not both at once.

These facts are well illustrated by the stereoscope. In looking through this instrument, there is set before each eye only a surface and not a solid figure. It is not the same surface which is set before both eyes, but a separate one before each. On these surfaces there is outlined a drawing of the same solid object, but the one surface presents it as seen from a point of view to the left, the other from a point of view to the right. The one figure represents the solid object as seen by the right eye, the other as seen by the left, when both are fixed on it. The result is the presentation, not of two superficial delineations, but of one solid object. The reason is, that when the two eyes are respectively fixed on corresponding parts of the two outlines, other parts of the field of view produce disparate impressions on the retina, just as they would do in looking at the same point of the actual object. The further they lie behind or before this point in the actual object, the more disparately situated are the impressions they produce, and the same is the case in looking through the stereoscope. The solid effect in the stereoscope is greatest when the eyes are allowed to move freely from one point to another. But it is also unmistakably present when the illumination by which the two pictures are seen is so transient as not to allow time for movements of the eyes. Of course as the eyes fixate one point after the other of the apparently solid object, retinal impressions which have been previously disparate come to affect corresponding points, and those which previously affected corresponding points become disparate. The appearance of solidity is more distinct and impressive, the fewer are the double images discerned. Old and practised experimenters, who concentrate their attention with the view of finding double images, become in time unable to obtain the stereoscopic effect. They see only a flat surface.