PART 5 A MONOGRAPH OF THE EXISTING CRINOIDS 637 



some of these arc collecting into ilistiiict balls of various sizes which lie between the 

 ectoderm and the entoderm. These groups of yolk granules remain a very character- 

 istic feature in the following developmental stages. 



There is some variation in the order of appearance of the ectoderm and entoderm. 

 In one case the nuclei of the ectodenn still form a simple nearly regular layer, though 

 the delimitation of the cells is well advanced, while the entoderm is fairly well along in 

 development; in another the nuclei of the ectoderm are much more numerous and are 

 not arranged in a simple layer, the delhnitation of the cells is less advanced, and the 

 entoderm is in a considerably j^ounger developmental stage, the cavity having only 

 just appeared. In another embryo of a stage corresponding e.xactly to the first of these 

 two there is a distinct indication of cilia in one place. 



The differentiation of the archenteron proceeds as m Antedon and Tropiometra. 

 A median constriction separates the archenteron into an upper or anterior section with 

 a very large lumen, the enterohydrocoele, and a lower or posterior section, the coelomic 

 sac, which divides into a right and left enterocoele vesicle. There is no trace of the 

 downward prolongations from the anterior vesicle, which in Antedon embrace the 

 narrow transverse canal that connects the two enterocoelic vesicles. Mortensen could 

 not make out the formation of the hydi'ocoele in all its details; but it is formed from the 

 anterior vesicle as in Antedon and Tropiometra. The parietal canal is formed in the usual 

 way as a constriction from tlie Iiydrocoele vesicle. 



The ectoderm consists of high epithelial cells, the nuclei of whicii lie chiefly in the 

 middle forming a more or less definite layer. The entoderm also is made up of cylin- 

 drical epithelial cells with the nuclei lying chiefly at the base and the whole interior of 

 the cells filled with yolk spherules. The enterocoelic vesicles show the same epithelial 

 structure. At the stage at which the formation of the archenteron begins mesenchyme 

 cells appear. Whether they originate from the entodermal epithelial cells or from the 

 nuclei originally lying scattered in the yolk mass cannot be made out; but from tlie fact 

 that they lie at first mainly at the upper end of the archenteron and do not fill the 

 blastocoele cavity, which has now become very large, the evidence is decidedly in favor 

 of their originating from the entodermal epithelium as in Antedon and in Tropiometra. 

 Eventually the blastocoele cavity becomes completely filled with mesenchyme cells. 

 From the first these are filled with yolk spherules, like all the other cells. But gradually 

 the yolk spherules collect in rounded groups which lie scattered irregularly, maiidy in 

 the mesenchyme. The nuclei of the ectoderm, entoderm and mesenchyme cells are 

 not distinctly different m size. 



Important structural changes now take place leading to the organization of the 

 fully developed larva. In its external appearance the embryo becomes more elongate. 

 The vestibulary invagination appears as a depression on the ventral side, and another 

 depression is formed at the anterior end of the embrj^o representing the suctorial disk. 

 The ciliated bands are developing, the nuclei being arranged in more or less distinct 

 groups corresponding to the bands. Glandular cells have developed in the ectodenn, 

 especially at the anterior end, but they may be nmnerous also in other places. The 

 nervous sj-stem has just begun to difi"erentiate. 



The entoderm shows a marked dOTerence from the condition in the preceding stage. 

 The lumen has now become ver3^ small, or is even completely obliterated. The cell 

 limits have mostly disappeared, and it is especiall}' important to notice that the cells 

 are not distinctly delimited from the lumen of the entoderm as they are in the preceding 



