cockle or mussel. But for the toughened surface representing the cockle's foot, we might regard the oyster as a lapsed form of some ancient Cephalopod.
The mantle secretes the shell, and in all bivalves it lies through its whole extent against the shell. Now, in all mollusks, the axis of the body is at first straight, and the body is bisymmetrical. If growth were arrested at an early stage, all mollusks would look alike, and, if the embryotic mantle were to secrete a shell, all these arrested growths would appear as miniature bivalves. They would be symmetrical. But circumstances determine shapes. The mollusk which, in maturity as well as infancy, lives in the open sea, will be exposed to like conditions on either side, and will retain its bilateral symmetry. A mollusk which lies on the sea-bottom will be exposed to unlike conditions, one side being buried in mud and the other bathed in water. As a shrub which grows against a wall loses its symmetry and becomes one-sided, so a young oyster, as soon as it leaves off its roving ways, and fixes its abode on the sea-mud, must begin to develop unsymmetrically. One side and one valve of the shell outgrow the other side and valve. In the Gryphgæa, an ancient species of oyster, this over-development of one side is carried further, and, while the right valve is small and flat, the left is deep and partially rolled up. In the Gasteropods, except Chiton, this one-sidedness is carried still further. One side outgrows the other so much that the body takes a spiral form, and one valve, secreted by one fold of the mantle, appears as a spiral shell, while the other valve, secreted by the aborted fold of the mantle, appears as an operculum—a little shelly disk known under the name of "eye-stone," In the snail this one-sided development is carried to the highest pitch of asymmetry. Overgrowth of the right side forces it into a spiral, and the right valve twists around with the body it incloses, while the left valve, which, in the marine Gasteropod, we had found reduced to an operculum, is here completely eliminated.
From the cloaked clavelina to the oyster, we were led, step by step, along successive modifications of the mantle. From the oyster to the snail we have passed, step by step, along successive stages of one-sided over-development. The facts have shown that a bivalve mollusk could not have descended from a univalve. As all mollusks in early life have the axis of the body straight, and the parts symmetrically arranged on either side, we may infer that bilateral symmetry characterized the remote ancestors of the molluscan type. Now, while a mollusk is bisymmetrical or nearly so, if the mantle secretes a shell it must be in in two parts, or, as in Chiton, in many parts. The snail is the last term of our series, and its successive stages of growth should indicate the path along which Nature has moved in the evolution of the unsymmetrical Gasteropod from a symmetrical, oysterlike bivalve (Fig. 4).
Lereboullet has made out the embryology of Limneus, a fresh-
