whom, since they have migrated to dark places, sight is useless? I think there is abundant evidence proving that it has so aided. An eye, if useless, must be much worse than useless to the animal possessing it. In its exposed position it is in constant danger of injury, and, from the delicate nature of the organ, harm to it more disastrously affects the whole organism than harm to an equal extent of any other part of the surface. Eyeless animals, or animals with small rudimentary eyes, therefore, are at an advantage in the dark, as compared to animals with fully-developed eyes. The latter would therefore tend to be eliminated, while the former would tend to survive and have offspring. Accordingly we find that certain cavern-inhabiting Crustacea, descended from stalk-eyed ancestors, have lost their eyes, but not the stalks on which the eyes were carried. The loss of the eyes cannot be set down to Cessation of Selection alone, for, since the stalks must have appeared later in the phylogeny than the eyes, the former are presumably the less stable structures, and therefore liable to disappear sooner on Cessation of Selection, whereas they have persisted. In like manner the disappearance of the eyes cannot be attributed entirely to Cessation of Use, for, if Cessation of Use is a cause of specific degeneration, it ought to have caused an equal or indeed greater retrogression of the less stable, stalks. Therefore in this case we can account for the persistence of the stalk, when the eye has disappeared, only by supposing that Natural Selection—reversed Natural Selection—has co-operated with Cessation of Selection (or Cessation of Use, according to the theory we hold) to remove the worse than useless eye, leaving the no more useful but much less harmful stalk persistent. Similarly we can explain the fact that the degenerate eyes of the proteus are covered with skin neither by the theory of Cessation of Selection nor by the theory
