[;7 History of a transient nervous apparatus in certain Ichthyopsida. 333 



Within the cord also there can be seen two such cells (gl. ce) 

 on each side. 



All these are ganglion-cells in course of development and they 

 form nascent elements of the future transient system. 



In another embryo (R. hatis, No. 100), of 5,5 mm, two gill-pouches 

 are present, the pronephros is in course of development, and the seg- 

 mental duct is only present in the region of, and in connection with, 

 the pronephros. This embryo differs little and in unimportant re- 

 spects from Nos. 134 and 135, but it is slightly "younger" than 

 either of these, and transient nervous elements have as yet no existence. 



The following embryo (R. hatis, No. 135, 6 mm) is in an interest- 

 ing phase. 60 trunk somites were counted. There is no sign of lens 

 formation, and, though the primary optic vesicles reach the epiblast, 

 there is no infolding to form a cup. There is a very slight depression 

 of epiblast in the region of the auditory organ. Traces of four gill- 

 pouches are present, but none are open to the exterior. 



The pronephros is in course of development, and the segmental 

 duct is formed for part of the length of the pronephros. A neuren- 

 teric canal is of course present. 



In the region of the pronephros the ganglionic foundations are 

 migrating from their position between the lips of the cord. 



Of the uppermost cells of the cord in this region some are taking 

 on ganglionic characters. These cells are large and rounded (Fig. 2 

 gl. c), the nucleus is also of some size, and it, like the cell itself, 

 does not stain as deeply as do the nuclei and cell-protoplasm of 

 neighbouring cells. Large refractile nucleoli, more than one in each 

 cell, also attract the notice. Some of these cells are, moreover, spinn- 

 ing out nerve-processes (Fig. 3 n.p). Still further back along the 

 trunk, beyond the region of the pronephros, there are even numerous 

 instances of this. These processes are, however, but in the very first 

 stages of their formation, and, as yet, the transient ganglionic appa- 

 ratus is hinted at rather than demonstrated. As already described 

 of Fig. 5, here also such developing ganglion-cells are recognisable 

 in both the cord and ganglionic foundations (Figs. 2 and 3 gl. c and 

 gl ce). In Fig. 4, which is also from the pronephric region, two 

 cells (w.gl.c) are figured at the apex of the myotome. These are 

 both already ganglionic, and form two wandering elements of the 

 transient system. 



Embryo No. 81 measured 8 mm. unfortunately no record of the 



