494 Cresswell Shearer 



essential respects. Except in several minor fcatures I can corroborate 

 most of the recent aceounts of the origin of the mesoderm. 



I bave been unable to find any trace of the bhistocoelic pore 

 mentioned by Ikeda as present at the end of segmentation. I have 

 also been unable to find the mesoderm arising* from lateral archen- 

 teric folds. According to Ikeda (14) this is one of the principal 

 sources of the origin of the mesoderm in the Japanese species. 



The blastulae are frequently very irregulär in shape from 

 mutuai pressure in the tentacles of the adult, individuai cells being 

 pushed out of place. Till the end of invagination no cells are to 

 be distinguished in the segmentation cavity. On this point the 

 majority of modern workers seem to agree, while earlier workers, 

 dependent on optical sections alone, usually figure mesenchyme 

 cells in the segmentation cavity during and after invagination. As 

 invagination takes place over the entire ventral surface of the blas- 

 tula the segmentation cavity is completely obliterated by the folding 

 up of the ventral upon the dorsal surface (pl. 31 figs. 1 — 4). Shortly 

 after this bilateral symmetry makes its appearance by the elongation 

 of the gastrula in one axis (fig. 6). The lips of.the blastopore 

 close up rapidly from behind forwards meeting in the median line. 

 Along this line of closure, over a short area where the lips of the 

 blastopore are meeting, is a region said to give rise to mesoderm (fig. 5). 

 I have examined this area repeatedly by means of sections without 

 bcing able to see the actual formation of mesoderm cells. I believe 

 that the active celi proliferation of this region has to do solely with 

 the closure of the blastopore and not with the origin of mesoblast. 

 Later however, there is an active formation of mesoderm on either 

 side of the line along which the blastopore has closed. This takes 

 place by individuai cells being pressed into the segmentation cavity 

 between the two layers from the endoderm (fig. 7). Towards the end of 

 invagination a certain number of mesoderm cells have made their 

 appearance from the ends of the invaginating endoderm cells. These 

 are especially numerous over that region of the endoderm which 

 will be later the pre-orai region. In the larvae from Faro these 

 cells at this stage form what seems a closed sac (fìgs. 26 and 27). 

 This is similar to the sac described by Oowles in Ä. arcJftfrcla (6). 

 This sac grows larger with the growth of the pre-orai region, so 

 that it Comes to have a horse-shoe shapod form, the two horns of 

 the shoe cxtending back into the trunk region, where they are formcd 

 of the mesoderm cells derived from division from the lateral walls 



