91 



certain of the indifferent cells of tlie margin participate in the forma- 

 tion. The only diflference is that at this period the notochord 

 remains connected with the entoderm, the mesoderm early loses its 

 connection A\ith the entoderm excej^t at the ]ilace of origin. By 

 this change there are interposed between the ectoderm and the 

 entoderm at the growing margin a median notochord, continuous 

 on each side with ^^'ings of mesoderm which extend all round the 

 margin. Jn point of origin the notochord and the mesoderm are 

 the same. This serves still further to emphasise the cavity above 

 referred to. It is a '^ mesoderm-free " area, and marks as it possibly 

 always does the region of the primitive gastrula. 



This will be still more plain from a consideration of the trans- 

 verse sections of a slightly older embryo (fig. 4). The figures are 

 drawn from sections (a) through the mesoderm-free area in front of 

 the brain, (6) immediately in front of the notochord, (c) and {d) 

 through the anterior and posterior ends of the as yet short noto- 

 chord. It will be at once apparent to those familiar with the sub- 

 ject that, apart from the svncytial condition of the entoderm and 

 the massiveness of the ectoderm, the sections bear a remarkable 

 resemblance to those obtained from similar stages of Selachii. The 

 median notochord is formed from a proliferation of entodermal 

 cells, syncytial though they be, but as has been seen from fig. 36 

 the structure is contributed to also from the marginal cells. On 

 each side of the notochord for a short distance — and altogether 

 independently of the marginal cells — a proliferating area is developed 

 in the syncytial entoderm, which gives rise to the enteron entoderm. 

 The notochord passes insensibly into the marginal mesoderm, 

 which, as has been seen, has a similar origin, viz., from the syncytial 

 entoderm and from the marginal cells. It cannot truly be said 

 here at all events that the mesoderm is divided into an axial and a 

 marginal. Both have the same origin in the margin, and so has the 

 notochord. The notochord may therefore be regarded as the axial 

 mesoderm. This is not a new suggestion, but it is the first time 

 it has been so clearly indicated from embryological work. If the 

 above explanation of the origin of the notochord be considered to 

 be sufficient, then it will no longer be necessary, for example, to 

 refer chorda cartilage to ectoderm or to entoderm. 



A layer, even though it is syncytial, which encloses the yolk, 

 furnishes entirely the embryonic entoderm, and yields to such a 

 large extent the notochord and mesoderm can only be the entoderm. 



