Fig. 317.—Brood-pouch of Thecidium mediterraneum. (After Lacaze-Duthiers.) Part of the wall of the pouch has been removed to show the clusters of larvae.

In Cistella the larva consists at first of two segments, but the anterior one divides into two, so that in the free swimming larva we find three segments, the hindermost somewhat longer and narrower than the others and destined to form the stalk. About the time of the appearance of the second segment four red eye-spots arise in the anterior segment, which tends to become constricted off from the others, and may now be termed the head. It gradually becomes somewhat umbrella-shaped, develops cilia all over its surface and a special ring of large cilia round its edge.

In the meantime the second or mantle segment has grown down and enveloped the stalk, and four bundles of setae have arisen from its edge. In this stage the larva leaves its mother’s shell and swims out into the world of water to look for a suitable place on which to settle down. This is the only stage in the life history of a Brachiopod when the animal is locomotor, and can serve to spread its species. The extreme minuteness of the larva and the short time it spends in this motile condition probably accounts for the fact that Brachiopods are extremely localised. Where they do occur they are found in great numbers, rocks being often almost covered with them, but they are not found over large areas. When viewed under a microscope the larvae seem to be moving with surprising rapidity, but judging from the analogy of other forms, it seems doubtful if they swim a yard in an hour.

Fig. 318.—Young larva of Cistella neapolitana, showing three segments, two eye-spots, and two bundles of setae. (After Kowalevsky.)

Fig. 319.—Full-grown larva of Cistella neapolitana, with umbrella-shaped head, ciliated. (After Kowalevsky.)

Frequently the larva stands on its head for some time, as if investigating the nature of the rocks on which it may settle; it is extremely contractile, turning its head from time to time, and seldom retaining the same outline for any length of time; the setae are protruded, and at times stick out in every direction; they are possibly defensive in function. When fully stretched out the larva is about ⅓ mm. long, but it frequently shortens its body to two-thirds of this length. The larvae are of a pinkish red colour, with eye-spots of ruby red. Their colour renders them difficult to discern when they are swimming over the red coralline rocks upon which they frequently settle. After swimming about for a few hours the larva fixes itself finally, apparently adhering by some secretion produced by the stalk segment. The folds of the second or body segment then turn forward over the head, and now form the ventral and dorsal mantle folds; these at once begin to secrete the shell on their outer surfaces. The head with its eye-spots must be to some extent absorbed, but what goes on within the mantle is not accurately known. The setae drop off and the tentacular arms begin to appear as a thickening on the dorsal lobe of the mantle. They are at first circular in outline. The various changes which the larva passes through are well illustrated by Morse for Terebratulina, which spawns at Eastport, Me., from April till August. The different stages are represented in outline in Fig. [320], taken from his paper.