Page:EB1911 - Volume 04.djvu/380

 regards the earlier stages. Segmentation is complete, a gastrula is formed, the blastopore closes, the archenteron gives off two coelomic sacs which, as far as is known, are unaffected by the superficial segmentation of the body that divides the larva into three segments.

The walls of these sacs give rise at an early stage to muscles which enable the parts of the larva to move actively on one another (fig. 29, B). About this stage the larvae leave the brood-pouch, which is a lateral or median cavity in the body of the female, and lead a free swimming life in the ocean. The anterior segment broadens and becomes umbrella-shaped; it has a powerful row of cilia round the rim and smaller cilia on the general surface. By the aid of these cilia the larva swims actively, but owing to its minute size it covers very little distance, and this probably accounts for the fact that where brachiopods occur there are, as a rule, a good many in one spot. The head bears four eye-spots, and it is continually testing the ground (fig. 29, A, C). The second segment grows downwards like a skirt surrounding the third segment, which is destined to form the stalk. It bears at its rim four bundles of very pronounced chaetae. After a certain time the larva fixes itself by its stalk to some stone or rock, and the skirt-like second segment turns forward over the head and forms the mantle. What goes on within the mantle is unknown, but presumably the head is absorbed. The chaetae drop off, and the lophophore is believed to arise from thickenings which appear in the dorsal mantle lobe. The Plankton Expedition brought back, and H. Simroth (Ergeb. Plankton Expedition, ii., 1897) has described, a few larval brachiopods of undetermined genera, two of which at least were pelagic, or at any rate taken far from the coast. These larvae, which resemble those described by Fritz Müller (Arch. Naturg., 1861–1862), have their mantle turned over their head and the larval shell well developed. No stalk has been seen by Simroth or Fritz Müller, but in other respects the larva resembles the stages in the development of Megathyris and Terebratulina which immediately precede fixation. The cirri or tentacles, of which three or four pairs are present, are capable of being protruded, and the minute larva swims by means of the ciliary action they produce. It can retract the tentacles, shut its shell, and sink to the bottom.

C. E. E. Beecher (Amer. Jour. Sci. ser. 3, xli. and xliv.) has classified with appropriate names the various stages through which Brachiopod larvae pass. The last stage, that in which the folds of the second segment are already reflected over the first, he calls the Typembryo. Either before or just after turning, the mantle develops a larval shell termed the protegulum, and when this is completed the larva is termed the Phylembryo. By this time the eyes have disappeared, the four bundles of chaetae have dropped off, and the lophophore has begun to appear as an outgrowth of the dorsal mantle lobe. The protegulum has been found in members of almost all the families of Brachiopod, and it is thought to occur throughout the group. It resembles the shell of the Cambrian genus Iphidea [Paterina], and the Phylembryo is frequently referred to as the Paterina stage. In some orders the Phylembryo is succeeded by an Obolella stage with a nearly circular outline, but this is not universal. The larva now assumes specific characters and is practically adult.

Classification.—Beecher’s division of the Brachiopoda into four orders is based largely on the character of the aperture through which the stalk or pedicle leaves the shell. To appreciate his diagnoses it is necessary to understand certain terms, which unfortunately are not used in the same sense by all authors. The triangular pedicle-opening seen in Orthis, &c., has been named by James Hall and J. M. Clarke the delthyrium. In some less primitive genera, e.g. Terebratula, that type of opening is found in the young stages only; later it becomes partly closed by two plates which grow out from the sides of the delthyrium. These plates are secreted by the ventral lobe of the mantle, and were named by von Buch in 1834 the “deltidium.” The form of the deltidium varies in different genera. The two plates may meet in the middle line, and leave only a small oval opening near the centre for the pedicle, as in Rhynchonella; or they may meet only near the base of the delthyrium forming the lower boundary of the circular pedicle-opening, as in Terebratula; or the right plate may remain quite distinct from the left plate, as in Terebratella. The pro-deltidium, a term introduced by Hall and Clarke, signifies a small embryonic plate originating on the dorsal side of the body. It subsequently becomes attached to the ventral valve, and develops into the pseudo-deltidium, in the Neotremata and the Protremata. The pseudo-deltidium (so named by Bronn in 1862) is a single plate which grows from the apex of the delthyrium downwards, and may completely close the aperture. The pseudo-deltidium is sometimes reabsorbed in the adult. In the Telotremata neither pro-deltidium nor pseudo-deltidium is known. In the Atremata the pro-deltidium does not become fixed to the ventral valve, and does not develop into a pseudo-deltidium. The American use of the term deltidium for the structure which Europeans call the pseudo-deltidium makes for confusion. The development of the brachial supports has been studied by Friele, Fischer and Oehlert. A summary of the results is given by Beecher (Trans. Connect. Acad. ix., 1893; reprinted in Studies in Evolution, 1901).

The orders Atremata and Neotremata are frequently grouped together, as the sub-class Inarticulata or Ecardines—the Tretenterata of Davidson—and the orders Protremata and Telotremata, as the Articulata or Testicardines— the Clistenterata of Davidson. The following scheme of classification is based on Beecher’s and Schubert’s. Recent families are printed in italic type.

Class I.

ORDER I. Atremata (Beecher).—Inarticulate Brachiopoda, with the pedicle passing out between the umbones, the opening being shared by both valves. Pro-deltidium attached to dorsal valves. FAMILIES.—

'''ORDER II. Neotremata''' (Beecher).—More or less circular, cone-shaped, inarticulate Brachiopoda. The pedicle passes out at right angles to the plane of junction of the valves of the shell; the opening is confined to the ventral valve, and may take the form of a slit, or may be closed by the development of a special plate called the listrium, or by a pseudo-deltidium. Pro-deltidium attached