DEVELOPMENT OF THE TWO MIDDLE GERM-LAYERS. 119 



appear in stronger relief. The process of separation is introduced 

 by the curving of the chordal plate, and its conversion into the 

 chordal groove (fig. 79 A ch). Inasmuch as it is continuous at its 

 edges with the parietal lamella of the middle germ-layer (mh^), there 

 arise in the roof of the ccelenteron the two small chordal folds, which 

 enclose between them the chordal groove. Tts free margins abut 

 directly upon the folded edge, where the visceral lamella of the 

 middle germ-layer (mk^) bends around into the entoderm (ik) to 

 produce the ccelenteric fold. 



In the next following stage (fig. 79 B) the thickened medullary 

 plate, consisting of long cylindrical cells, becomes distinctly marked 

 oif from the now still smaller cubical elements of the ectoderm. 

 Meanwhile the middle germ-layer begins to detach itself from its 

 previous connections in the vicinity of the place of evagination ; the 

 parietal lamella becomes separated from the fundament of the 

 chorda, the visceral lamella from the entoderm, and thereupon then- 

 detached edges become fused to each other. By means of this pro- 

 cess the fundament of the body-sac, or of the middle germ-layer, 

 becomes closed on all sides, and is separated from the other 

 germ-layers. At the same time the entoderm (iki) and the funda- 

 ment of the chorda {oh) have come into contact along their free 

 margins, so that the chorda appears like a thickening of the ento- 

 derm, and for a time shares in bounding the intestinal cavity on the 

 dorsal side. This is changed by a second process of detachment. 



The fundament of the chorda, now converted into a solid rod, 

 is gradually excluded from participation in Hning the intestine 

 (fig. 79 C), by the fact that the halves of the entoderm (ik), composed 

 of large yolk-cells, grow toward each other underneath it, and fuse 

 in a median raphe. 



The closure of the permanent intestine on the dorsal side, the con- 

 stricting off of the two hody-sacs from the inner germ-layer, and the 

 origin oj the chorda dorsalis are therefore in Amphibia, as in Amphi- 

 oxus, processes which are most intimately related with one another. 

 Here, too, constricting off oJ the parts m,entioned begins at the head-end 

 of the embryo, ami advances slowly toward the posterior end, where 

 there exists for a long time a zone of growth, by means of which the 

 increase in the length of the body is effected. Soon after this, the 

 moment arrives when in the embryos of Triton the body-cavity 

 becomes visible. For after the detachment of the organs previously 

 mentioned is completed, the two middle germ-layers at the head-end 

 of the body, and on both sides of the chorda, separate from each 



