It is said, however, that _Chologaster agassizii_, which is not blind, lives in the underground streams of Kentucky and Tennessee, but I think it is open to doubt whether it is a species entirely confined to darkness. Another point which Loeb omits to mention is the absence of pigment in cave animals, especially Vertebrates such as _Amblyopsis_ and _Proteus_. If absence of light is not the cause of blindness in these cases, how is it that the blindness is always associated with absence of pigment, since we know that the latter in Fishes and Amphibia is due to the absence of light? It has been shown that _Proteus_ when kept in the light develops some amount of pigment, although it does not become pigmented to the same degree as ordinary Amphibia. We have here, I think, an example of the essential difference between mutations and somatic modifications. Absence of the gametic factor or factors for pigmentation results in albinism, and no amount of exposure to light produces pigmentation in albinos, _e.g._ albino Axolotls which are well known in captivity. Absence of light, on the other hand, prevents the development of pigment. The question therefore is whether the somatic modification is inherited. The fact that _Proteus_ does not rapidly become as deeply coloured when exposed to light as ordinary Amphibia shows that the gametic factors for pigmentation have been modified as well as the somatic tissues. Loeb attributes the blindness of cave fishes to a disturbance in the circulation and mutation of the eyes originally occurring as a mutation. But how could an explanation of this kind be applied to the case of _Anableps tetrophthalmus_, in which each eye is divided by a partition of the cornea and lens into an upper half adapted for vision in air and a lower half for vision in water? This fish lives in the smooth water of estuaries in Central America, and swims habitually with the horizontal partition of the lens level with the surface of the water. It is impossible to understand in this case, firstly, how a mutation could cause the eyes to be divided and doubly adapted to two different optic conditions, and, secondly, how at the same time a convenient 'tropism' should occur which caused the animal to swim with its eyes half in and half out of water. Are we to suppose that the upper half of the body or eye had a positive heliotropism and the lower half a negative heliotropism? The fact is that the fish swims at the surface in order to watch for and feed on floating particles. The tropism concerned is the food tropism, but what is gained by calling the search for food common to all active animals a tropism, and how is the search for food before the food is perceptible to the senses, before it can act as a stimulus on a food-sensitive substance in the body, to be compared to a tropism at all?