exact shape of the valves, more especially of the terga. The adductor muscle is always attached to a point not far from the middle of the scuta, and it generally has a pit for its attachment. In several genera, namely, Pæcilasma, Dichelaspis, Conchoderma, and Alepas, the scuta show a tendency to be bilobed or trilobed. The valves are placed either at some distance from each other, or close together; but their growing margins very rarely overlap each other, though this is sometimes the case with their upper, free, tile-like apices; in a few species the scuta and terga are articulated together, or united by a fold. The membrane connecting the valves, where they do not touch each other, is like that forming the peduncle, and is sometimes brilliantly coloured crimson-red; generally, it appears blueish-gray, from the corium being seen through. Small pointed spines, connected with the underlying corium by tubuli, are not unfrequently articulated on this membrane: the tubuli, however, are often present where there are no spines. To allow of the growth of the capitulum, the membrane between the valves splits at each period of exuviation, when a new strip of membrane is formed beneath, connected on each side with a fresh layer of shell,--the old and outer slips of membrane disintegrating and disappearing: when there are many valves, the line of splitting is singularly complicated. This membrane consists of chitine,[13] and is composed of numerous fine laminæ. After the valves have been placed in acid, a residue, very different in bulk in different genera, is left, also composed of successive laminæ of chitine. It appears to me that each single lamina of calcified chitine, composing the shell, must once have been continuous with a non-calcified lamina in the membrane connecting the several valves: at the line where this change in calcification supervenes, the chitine generally assumes some colour, and becomes much harder and more persistent; and as the whole valve is formed of component laminæ thus edged (the once continuous laminæ of non-calcified chitine connecting the valves, having disintegrated and disappeared) the surfaces of the valves are generally left covered by a persistent membrane, constituted of these edgings: this membrane has been called the epidermis. In some genera, as in Lepas, this so-called epidermis is seldom preserved, excepting on the last zone of growth: in Scalpellum and Pollicipes it usually covers the whole valves. It appears to me that the laminæ of chitine, and of calcified chitine composing the valves, are both formed not by secretion, but by the metamorphosis of an outer layer of corium into these substances.