This begins by the fish (owing to the increasing depth of the body) being unable to maintain the vertical position, so that it falls on one side.

It then twists the lower eye as much as possible towards the upper side; and, the whole bony structure of the head being at this time soft and flexible, the constant repetition of this effort causes the eye gradually to move round the head till it comes to the upper side.

Now if we suppose this process, which in the young is completed in a few days or weeks, to have been spread over thousands of generations during the development of these fish, those usually surviving whose eyes retained more and more of the position into which the young fish tried to twist them, the change becomes intelligible; though it still remains one of the most extraordinary cases of degeneration, by which symmetry--which is so universal a characteristic of the higher animals--is lost, in order that the creature may be adapted to a new mode of life, whereby it is enabled the better to escape danger and continue its existence.
The most difficult case of all, that of the eye--the thought of which even to the last, Mr.Darwin says, "gave him a cold shiver"-- is nevertheless shown to be not unintelligible; granting of course the sensitiveness to light of some forms of nervous tissue.

For he shows that there are, in several of the lower animals, rudiments of eyes, consisting merely of pigment cells covered with a translucent skin, which may possibly serve to distinguish light from darkness, but nothing more.

Then we have an optic nerve and pigment cells; then we find a hollow filled with gelatinous substance of a convex form--the first rudiment of a lens.