410 JOHN BEARD, 



extending backwards, it finally stretches over the region of 25—26 

 somites. 



By the time all the gill-clefts are formed this region of the cord 

 has a tesselated roof of ganglion-cells. 



In a single transverse section from two to eight of these ganglion - 

 cells may be present. In Raja batis the average number in the region 

 of a mesoblastic somite is about 20, and the total of those on the 

 cord cannot be less than 500—600. Most, if not all, these central 

 ganghon-cells are multipolar, and they lie in a close meshwork of 

 their processes. The crossing of fibres from side to side is often 

 very apparent in transverse sections. The nerve-processes of these cells 

 are described further on. 



The portion of the apparatus at first developed is subsequently 

 somewhat reinforced by cells arising further back, even to the tail-end. 



An explanation is offered of the feebleness of this posterior 

 portion, and it is regarded as due to a carrying-back of parts of a 

 larva (represented by the blastoderm) into a region far beyond the 

 blastodermic limits, owing to the raising-up of the embryonic body 

 from the blastoderm and its growth backwards. 



Peripheral ganglion-cells of the transient system. 

 — This title is used purely in a descriptive sense to classify all those 

 gangUon-cells of the transient system, which do not lie at the top of 

 the spinal cord, but are met with either in the mesoderm, or in the 

 myotomes, or jnst underneath the epiblast. 



Going back to sections of early embryos the description of certain 

 cells among the wandering cells of the spinal ganglia, as previously 

 mentioned, is taken up. 



Some of the cells upon and among the cells of the future spinal 

 ganglia are seen to be becoming ganglionic. They are also in course 

 of migration into the mesoderm, and the goal, towards which they 

 appear to strive, is the apex of the myotome. So manifold are the 

 appearances presented by these wandering ganglion-cells, that it is 

 difficult to give an account of them without describing a great number 

 of special cases. One frequent feature is a tendency on the part of 

 the wandering cell to spin nerve- processes in the course of its mi- 

 gration. The processes are referred to again in treating of the nerves 

 of the system. Many of the wandering ganglion-cells become applied 

 to, or mixed-up with, nerves of the transient system. Another favourite 

 resting-place appears to be the epithelium of the myotome-apex. 

 Sometimes they form such regular members of this epithelium that 

 one is tempted to a belief in their origin in loco. 



