=Eyes of Onchidium.=--Many species of _Onchidium_, a naked land pulmonate which creeps on rocks near high-water mark, are provided with dorsal eyes of various degrees of organisation, and in numbers varying up to nearly one hundred. The tropical _Onchidium_ are the prey of a fish (_Periophthalmus_) which skips along the beach by the aid of its large ventral fins, and feeds principally on insects and _Onchidium_. Karl Semper suggests[290] that the eyes are of service to _Onchidium_ as enabling it to apprehend the shadow of the approaching _Periophthalmus_, and defend itself by suddenly contracting certain glands on the skin and expressing a liquid secretion which flies into the air like shot and frightens the _Periophthalmus_ away. This theory for it is no more than theory--may or may not be true, but it is remarkable that _Onchidium_ with dorsal eyes have precisely the same geographical distribution as _Periophthalmus_, and that where no _Periophthalmus_ exists, _e.g._ on our own S.W. coasts, the _Onchidium_ are entirely destitute of dorsal eyes. In those species of _Onchidium_ which have no dorsal eyes, the latter are on the tips of the tentacles, as in _Helix_. The eyes are developed _on_ the head, and afterwards ascend with the growth of the ommatophores, while in _Helix_ the ommatophores are formed first, and the eyes developed upon them.[291] =Dorsal Eyes in the Chitonidae.=--The remarkable discoveries of Moseley with regard to the dorsal eyes of _Chiton_ were first published in 1884.[292] He happened to notice, while examining a specimen of _Schizochiton incisus_, a number of minute black dots on the outer surface of the shell, which appeared to refract light as if composed of glass or crystal. These ‘eyes,’ in all the species of _Chiton_ yet examined, are restricted to the outer surface of the exposed area of the shell, never being on the laminae of insertion or on the girdle. In certain sub-genera of _Chiton_ the eyes are scattered irregularly over the surface, in others they are arranged symmetrically in rows diverging from the apex of each plate, but in old specimens the eyes towards the apices are generally rubbed off by erosion or abrasion. Moseley regarded the occurrence of scattered eyes as indicating an original stage of development, when the eyes were at first disposed irregularly all over the surface of the shell; the gathering into regular rows showing a later stage. [Illustration: FIG. 92.--Dorsal eyes of _Chitonidae_, showing the various forms of arrangement in the first and fourth valves of 1, 1_a_, _Acanthopleura spinigera_ Sowb., E. Indies, × 2; 2, 2_a_, _Tonicia suezensis_ Reeve, Suez, × 3; 3, 3_a_, _Acanthopleura granulata_ Gmel., W. Indies, × 2; 4, 4_a_, _Tonicia lineolata_ Fremb., Chili, × 2. From specimens in the Museum of Zoology, Cambridge.]