DEVELOPMENT OF THE PRIMITIVE SEGMENTS. 165 



musculature of the body, but from the cavities of the latter originates 

 the real unsegmented body-cavity, since the partitions which at 

 first separate them become thinner, break through, and finally 

 disappear. 



Similar processes take place in a somewhat modified manner in the 

 case of the remaining Vertebrates. 



In the Tritons the middle germ-layer (fig. 105 A) becomes 

 thickened on both sides of the chorda (ch) and of the fundament of 

 the central nervous system (ra), which is not yet closed into a tube r 

 and at the same time there appears a cavity (ush) in its thickened 

 part, caused by the separation of the visceral and parietal lamellae. 

 The thickening is not produced by an increase in the number 

 of the layers of cells, but simply by the fact that the cells 

 increase in height and grow out into long cylinders, which are 

 arranged around the cavity like an epithelium. We distinguish 

 these thickened parts of the middle germ-layer, which lie on either 

 side of the chorda and the nervous system, as the primitive-segment 

 plates, from the lateral parts, or the lateral plates. In the territory 

 of the latter the cells are lower, and ordinarily there is no distinctly 

 marked cavity between visceral and parietal layer. 



Whereas in Amphioxus the process of forming somites extends 

 itself over the whole of the middle germ-layer, in the case of the 

 Amphibians, and likewise all the re- 

 maining Vertebrates, it affects only 

 the part which is next to the chorda 

 and the neural tube, leaving the lateral 

 plates, on the contrary, untouched. 

 The segmentation begins at the head- 

 end, and prbceeds slowly toward the 

 blastopore ; it is accomplished by fold- 

 ing and constricting off. The epithelial 

 lamella next to the neural tube and Fig. ioe. Frontal section through 

 the chorda, being composed of cylin- 

 drical cells, is raised up into small ments. 

 transverse folds, which, separated from 

 each other by intervals of uniform size, with their cavities 

 grow into the cavity of the primitive- 

 segment plate, and give rise to small sacs lying one behind the other 

 (fig. 106). 



Soon afterwards each little sac is constricted off from the lateral 

 plates (fig. 105 A and ). Consequently one now meets, both in 



