I.

THE RELATIONS BETWEEN MESOBLAST AND ENTOBLAST IN ANNELIDS AND MOLLUSKS.

In Nercis, as in the typical development of other annelids and of gasteropods and lamellibranchs, the mesoblast-bands are derived from the posterior cell of the fourth quartet of "micromeres. This cell, now generally known as the second somatoblast, divides into two symmetrical halves which have been usually designated as the "primary mesoblasts;" and from them, by a series of slightly unequal successive divisions, arise the mesoblast-bands which extend forward in the cleavage-cavity at the sides of the embryo. Before giving rise to the mesoblast-bands, however, the "primary mesoblasts" bud forth the small cells already referred to, at or near the surface directly behind the two posterior macromeres "C" and "D." At least six, and probably not less than ten, of these cells are formed, the primary mesoblasts meanwhile sinking below the surface and becoming quite covered by ectoblast-cells which advance from the sides. and from behind. The small cells first formed lie at the surface, wedged in between the "primary mesoblasts" and the macromeres (Fig. 1, D, e; Fig. 2, B,). Those formed later lie below the surface, owing to a change in the plane of division (Fig. 3, 4). The small cells, which are very conspicuous in sections by reason of their intensely chromatic, closely reticulated nuclei, thus become arranged in a thin plate extending inwards from the surface between the primary mesoblasts and the two posterior macromeres (Fig. 3, B). After the formation of the small cells the divisions of the primary mesoblasts suddenly change both in form and direction, the plane of division being now nearly or quite at right angles to the former (i. e., approximately parallel to the sagittal plane of the embryo) and the cells thus produced being nearly as large as the primary

1 Nereis is somewhat exceptional in the fact that the other three cells of the fourth quartet are suppressed. In Aricia, Polymnia, Spio, Pysgmobranchus, Hydroides, Polygordius (all of which I have examined), and in some others, the fourth quartet, is complete, and in the first two forms named, a fifth quartet of (entoblastic) micromeres is formed before the invagination (Cf. Fig. 2, A).

FIG. 1.1

Early embryos of Aricia (A, C) and Nereis (B, D) in sagittal section (A, B, C, optical, D, actual). Showing the formation of small posterior entoblasts (e) between M and D.

A, B, D, b', cells of the entoblast-plate (cf. Fig. 2); M, the "primary mesoblast ;" m, mesoblast-band; X, the first somatoblast or its derivatives, forming the somatic plate.

mesoblasts. Thus are formed the mesoblast-bands which form together a V-shaped mass of cells lying between the macromeres and the overlying ectoblast. Near the middle line the two halves of the V are often slightly separated; and into the space

All the figures are from camera drawings, made from preparations unless otherwise stated. Optical sections have been fully confirmed by actual.

X

thus formed some of the small cells usually extend, appearing in sections in the sharpest contrast both to the large rounded mesoblast-cells and to those of the lateral ectoblast (Fig. 3, C). From this point the mesoblast-bands extend towards the sides and ultimately curve upwards (forwards with respect to the adult long axis) at the sides of the embryo.1

FIG. 2. Corresponding surface views, from the lower pole, of early embryos of Aricia (A) and Nereis (B); the limit of the ectoblast, i. e., the lip of the blastospore, is shown by the heavy line. A shows the single pair of vestigial entoblasts (e, e) of Aricia lying in front of the primary mesoblasts which are dividing to form the mesoblast-bands (cf. Fig. 1, C, which shows the same specimen in sagittal section). B shows two pairs of superficial entoblasts, lying behind the macromere D, and the spindles of a deeper budding of the "primary mesoblasts" (cf. Fig. 3, A, for section of this stage).

A, B, C, D, the four basal entoblasts or macromeres; a-, the fourth quartet of "micromeres (entomeres); a3-d5, the fifth quartet (entomeres); c3-d3, derivatives of the third quartet (ectomeres); M, M, the primary mesoblasts (shaded in B).

Up to this point the account here given is substantially the same as that contained in my earlier paper on Nereis. Regard

1 In Aricia the mesoblast-bands are formed much earlier, while the primary mesoblasts still lie at the surface (Fig. 1, C); and they lie at first side by side, nearly parallel to each other, extending upwards behind the entoblast-plate (Fig. 7). In both these respects Aricia is somewhat similar to Lumbricus (Cf. Wilson, Embryology of the Earthworm, Fig. 30: Journ. Morph., 1889).

3

ing the fate of the small cells, however, my first account was wide of the mark; for I believed that they migrated into the interior and spread out upon the walls of the archenteron to form a part of the splanchnic mesoblast.1 - I accordingly called the small cells "secondary mesoblast" and applied the same term to the rudimentary cells of Aricia and Spio. Later studies. by several observers seemed to confirm this conclusion. Lillie found in Unio a single pair of small superficial cells, budded forth from the "primary mesoblasts" exactly as in Aricia or Nereis, but relatively larger, which he likewise believed to wander into the cleavage-cavity to form a part of the mesoblast.2 Heymons found in Umbrella two pairs of corresponding but still larger cells, which he, too, apparently traced into the mesoblast. Mead found that a corresponding pair of minute cells, in Amphitrite are carried in at the tips of the mesoblast-bands ;* while Holmes still more recently states that in Planorbis they enter the segmentation cavity. Wierzejski's recent observations. on Physa, though differing from the foregoing in some important details, agree in referring the small cells, of which several pairs are formed, to the mesoblast. With such an array of confirmatory evidence my original conclusion seemed to be strongly supported. Conklin, however, in his remarkable paper on Crepidula, reached a wholly different result, finding in that gasteropod that cells which probably correspond with the small cells of Nereis, give rise to the posterior part of the archenteron. In regard to Nercis, I have long suspected that my original account of the fate of the small cells was erroneous. A renewed examination of the matter has left no doubt that such was the case, and gives the strongest ground for the conclusion. that, like the corresponding cells in Crepidula, they enter into the formation of the archenteron. The evidence for this conclusion is as follows:

In my earlier paper on Nercis I overlooked the fact that, besides the small cells derived from the "primary mesoblasts,"

other closely similar cells are formed, just in front of them by budding from the macromeres. These cells agree closely with those derived from the "primary mesoblasts" both in size and in the close reticulation and intensely chromatic character of

FIG. 3, NEREIS. Sections of successive stages in the formation of the entoblast-plug and mesoblast-bands in embryos of Nereis (actual sections, Flemming's fluid; Cis tranverse, the others sagittal). Lettering as before. A shows a deep budding of M (cf. Fig. 2, B); B, later stage showing group of small cells (e) derived from M; C, still later stage, nearly transverse, showing the mesoblast-bands (m, m) and the group of small cells (e) below; D, budding of the posterior macromere, D; E, recession of the entoblast-nuclei; F, first appearance of the pigment in the small cells.