DEVELOPMENT OP THE TWO MIDDLE GEEM-LAYEES. 141 



of the middle germ-layers detaoli themselves from the adjacent cell- 

 layers at the sides where their ingrowth took place, and at the same 

 time the halves of the permanent entoderm grow together, whereby 

 the dorsal closure of the intestine is effected. 



In view of these facts there is only one explanation at which we 

 can arrive. If it is certain that the middle germ-layers do not 

 arise by a fission in loco from either of the primary germ-layers, 

 then their gradual spreading out from a definite region of the germ 

 can result only from an ingrowth of cells, which occurs from those 

 places where a connection with other cell-layers has been demon- 

 strated. The middle germ-layers draw the principal material for 

 their growth from cells which, at the blastopore or at the primitive 

 groove, migrate between the two primary germ-layers. 



But this immigration of cells may be interpreted as a process of 

 infolding of the primary germ-layers, as in the case of Amphioxus. 

 In the method of the infolding there exists, it is true, one very 

 striking and apparently important difierence between Amphioxus 

 and the remaining Vertebrates. In Amphioxus the middle germ- 

 layer arises as a hollow sac, by means of the folding of the inner 

 germ-layer — in the remaining Vertebrates as a soUd mass of cells. 

 This undeniable difierence is, however, easily explained in the 

 following manner : In the solid fundaments of the middle germ- 

 layer a cavity is wanting, because the cellular walls of the sac are 

 from the beginning firmly pressed together, in consequence of the 

 yolk-mass which fills the coelenteron. In addition to other striking 

 agreements with the conditions in Amphioxus lanceolatus, there are 

 three points of view which in particular commend this interpretation : — 



(1) In all vertebrated animals there early arises in the middle 

 germ-layer a fissure, which is surrounded by cells, often cubical or 

 cyhndrical, having an epithelial arrangement. The parietal and 

 visceral layers then take the form of epithehal lamellse, as is to be 

 seen in an especially striking manner in the case of the Selachii at 

 a very early stage of development. (2) From these epithelial layers 

 there arise in the adult genuine epithehal membranes, Uke the 

 ciliated peritoneal epithelium of many Vertebrates, and, in addition, 

 glands that in many respects resemble the glands derived from 

 epithelial membranes [of the other germ-layers] (kidney, testis, 

 ovary). (3) The objection that the middle germ-layer of Verte- 

 brates arises as a single cell-mass, and therefore cannot be equi- 

 valent to two layers of epithelium, loses its weight with every one 

 who knows the numerous analogous phenomena of development 



