272 THE CHICK. 



somites are alone present (cf. Tig. 110), and sometimes even 

 prior to the definite formation of any of tlie somites. 



In the anterior part of the spinal cord the neural ridges 

 appear, just as in the brain, as cellular proliferations from the re- 

 entering angles between the external epiblast and the lips of the 

 neural plate, which latter have already grown in towards each 

 other a certain distance. The neural ridges of the spinal cord 

 are directly continuous with those of the brain, and from them 

 the ganglia ofthe spinal nerves are derived as paired outgrowths. 

 The spinal ganglia diifer, however, from the cranial ganglia in 

 not acquiring any distal connection with sensory patches of the 

 epiblast, and in being from th.e first much more deeply situated, 

 growing downwards close alongside the spinal cord, between 

 this and the muscle-plates (Fig. 124, ne). 



During the third day, the spinal ganglia acquire their definite 

 attachments to the sides of the spinal cord, these being eifected by 

 the outgrowth of nerve fibres from the inner sides of the ganglia 

 into the cord. The parts of the ganglia above, or dorsal to, 

 the points of attachment persist for some time as small pointed 

 processes, but soon become inconspicuous, and are finally absorbed 

 into the ganglia. 



The spinal ganglia are of considerable width, more than 

 half the width ofthe somites to which they belong (Fig. 115, ne), 

 so that the intervals between successive ganglia are distinctly 

 less than the width of the ganglia themselves. The ganglia lie, 

 from the first, opposite the anterior parts of the somites to which 

 they belong. 



The ganglia of the anterior part of the body are connected 

 together by short commissural bands (Fig. 115, nm), situated at 

 the same level as the attachments ofthe ganglia to the spinal cord ; 

 and the most anterior spinal ganglion, as already noticed, is con- 

 nected with the hindmost rodt of the pneumogastric nerve by a 

 similar but much longer conmiissure (Fig. 115, x"). These com- 

 missures appear to be formed from the parts of the originally 

 continuous neural ridge which are left between the successive 

 ganglion outgrowths ; they are well developed and conspicuous 

 structures on the fourth and fifth days, but after the latter date 

 are difficult to detect. 



The spinal nerves of the hinder part of the cord develop in 

 slightly different fashion to those ofthe anterior part. They appear 



