526 THE HUMAN EMBRYO. 



neural ridges, in the angles between the external epiblast and 

 the neural plate. On closure of the neural canal, the neural 

 ridges of the two sides become continuous with each other in the 

 median plane, to form the neural crest. The neural crest sepa- 

 rates from the external epiblast, but remains in close contact 

 with the spinal cord, forming a mass of spherical cells, wedged 

 in like a keystone between the dorsal edges of the neural plate. 



As the edges of the neural plate grow in towards each other, 

 to complete the dorsal wall of the spinal cord, the neural crest 

 is gradually squeezed out from between them, and its median 

 part thins away and disappears. From the lateral edges of the 

 neural crest outgrowths arise, which form the rudiments of the 

 spinal ganglia : these are at first exceedingly slender. 



The immediately succeeding stages in the development of 

 the ganglia have not been followed satisfactorily in human 

 embryos. About the middle of the fourth week the ganglia 

 have attained considerable size, and neuroblasts are present in 

 them in large numbers. These neuroblasts differ from those of 

 the spinal cord in being bipolar in place of unipolar, each 

 neuroblast giving off two processes in opposite directions, in- 

 wards and outwards respectively. The inwardly directed pro- 

 cesses grow from the ganglion into the spinal cord, and give 

 rise to the dorsal or sensory root of the nerve (Fig. 226, ND) ; 

 while the outwardly directed processes give rise to the sensory 

 portion of the trunk of the nerve. It is stated that all the cells 

 of a spinal ganglion send nerve processes into the spinal cord, 

 but it is not yet certain whether all the fibres of a dorsal root 

 are directly connected with ganglion cells. 



The later stages of development of the spinal nerves need 

 not be described in detail. The neuroblasts give rise directly 

 to the nerve cells of the cord and ganglia, each neuroblast, in 

 the later stages, giving off processes which come into close 

 relation with those of adjacent cells, but apparently do not anas- 

 tomose with these. Each nerve fibre arises in the first instance 

 as a process of a single cell or neuroblast, but it is not quite 

 clear in what mode its further growth is effected. His and 

 others maintain that it is simply by a continuation of the 

 process by which it first arose, and that the axis cylinder 

 throughout its whole length is to be regarded as a direct 

 prolongation of the body of the nerve cell from which it arises. 



