pletely reversed. It may indeed be said that the case of a a seen in front may be reconciled with the law of direction. For, if the combined images be referred to

the point of optic convergence A, as in the usual mode of representation, then each eye sees its own object in its true direction, but only mistakes its distance. To

m

FIG. 144.

aa

m

E

this I would answer that each eye does indeed give the true direction, as is quickly shown by shutting one of them, but the two eyes together do not. Each sees its

own object in the true direction, but the binocular observer sees their combination in a wrong direction. In the case of the double images m and m' (Fig. 144) of the object M (Fig. 143), it is still more difficult to explain their apparent position by the law of direction.

A curious Corollary. It is seen that, under all circumstances, if the median visual plane coincides with the median plane of the head, whatever be the position of the optic axes, objects in the visual lines are moved to the front and seen there. Now the same would be true if our eyes were turned directly outward right and

left.

There can be no doubt that if we could turn our eyes directly outward, or if our eyes, retaining their present organization and properties in regard to corresponding points, were transferred to the sides of the head with their axes straight right and left—i. e., making an angle of 180° with each other-images of objects in the direction of these axes, and therefore directly right and left, would be moved round 90° each, and combined and seen directly in front. This seems an extraordinary result, but it is a necessary consequence of the law of corresponding points. The retinal images of the two objects are on corresponding

points, viz., on the central spots; therefore, by the law of corresponding points, they must be seen as one. But where else can this take place but in front? The accompanying figures are a diagrammatic representation of these facts, Fig. 145 being the supposed condition of things, and Fig. 146 the visual result. After the frequent explanations of similar figures, a bare inspection will be sufficient.

It is needless to say that this is a purely hypothetical case. If any 'animals have their eyes so placed -i. e., on the sides of the head, and therefore optic axes like Fig. 145—they can not have corresponding points nor binocular vision. But of this we will speak further in the next chapter.

CHAPTER VI.

COMPARATIVE PHYSIOLOGY OF BINOCULAR VISION.

As we can not enter into the consciousness of animals, nor communicate intelligibly with them in regard to their visual experiences, we can only judge of these by the structure of their eyes. Three points of structure are important in this regard, viz., the optic chiasm, the position of the optic axes, and the presence or absence of a fovea.

Optic Chiasm. It will be remembered that in man, and also probably in most vertebrates, the optic roots, after leaving the brain, converge and unite to form the chiasm, and then again diverge as the optic nerves enter the eye sockets, pierce the eye, and spread out to form the retina. Furthermore, that in the chiasm the fibers of the roots partly cross over to form the fibers of the optic nerve on the other side, and partly do not cross over, but go to form the fibers of the optic nerve on the same side. This is shown diagrammatically in Fig. 41, page 119. Therefore each root supplies both optic nerves, and therefore both eyes, and conversely each eye is supplied by both roots and both sides of the brain. Still further, it is probable that the fibers of each root supply corresponding halves of the two eyes. There seems to be no doubt, therefore, that the optic chiasm, and especially this peculiar partial crossing of

the fibers, is in some way—imperfectly understood-intimately connected with the use of the two eyes as one instrument-i. e., with binocular vision. with binocular vision. I said especially the peculiar partial crossing, because by this arrangement each side of the brain controls both eyes. The bodily crossing over of fibers would not have this effect, for then each side of the brain would supply the opposite eye.

Now the optic chiasm, with its peculiar partial crossing of fibers, is probably present in all mammals and birds, and possibly in reptiles and amphibians. These, therefore, probably have, in a greater or less degree, perhaps imperfectly, the phenomena of binocular vision. But in fishes the fibers of the optic roots seem to cross bodily over to form the optic nerve on the other side. There is therefore in them no true chiasm, and therefore no true consensual movement of the two eyes and no binocular vision. We shall find other reasons for coming to this conclusion presently.

Nothing at all resembling an optic chiasm is found in any invertebrate. It is characteristic of vertebrates. No invertebrate enjoys the phenomena of binocular vision.

Position of the Optic Axes.-In man the axes of the eye-sockets diverge about 25° from one another, or about 12° each from the median plane of the head. In these slightly divergent sockets the eyeballs are so placed that their optic axes are parallel in a natural or passive state. This is evidently the most favorable position for easy convergence of the axes on an object at any distance, and therefore for binocular vision. A less divergence of the sockets, though still more favorable for convergence on a very near object, would produce too small an interocular base for accurate binocu