36 ILLINOIS BIOLOGICAL MONOGRAPHS {156 



The ventral cord arises as a thickening in the hypoderm, as has already 

 been shown, but later becomes separated from it, passing inward even 

 beyond the muscle layer and remaining connected with the hypoderm 

 only by a single row of cells (Figs. 58, 105). 



The cells that later make up the nerve cord at first appear as two rows 

 of larger cells in the hypoderm (Fig. 56) corresponding to the two rows of 

 nuclei on the ventral side of the larva. Even in later stages these two 

 rows of cells remain clearly distinguishable altho they crowd each other 

 so that they come to lie alternately one behind the other and do not usually 

 show in a single section. Between these two rows of cells and on each 

 side of them appear very early in development three longitudinal fiber tracts 

 (Figs. 56, 58), which are the rudiments of the three main fiber tracts of the 

 nerve cord. Nerve cells later appear under these fibre tracts and on the two 

 sides of them (Fig. 88), separating the tracts entirely from the rest of the 

 hypoderm. By the growth of the cells under the median tract, that is 

 pushed out beyond the two lateral tracts, and the division between the two 

 rows of cells becomes nearly obliterated. The cord, after separating from 

 the hypoderm, has in cross section the shape of a loop or fan with rounded 

 corners, the cells forming the base and projecting far into the interior. 



It has been impossible to determine the structure of the smaller cells. 

 The larger cells, where their structure could be made out, have been found 

 to be bipolar, giving off one fibre to the longitudinal tract and one to the 

 dorsal border of the cord (Fig. 139). The longitudinal fibres as well as the 

 radiating fibres stain very deeply with iron hematoxylin (Figs. 105, 106, 114) 

 but the structures shown throw little light on the physiology of the nerve 

 cord. The longitudinal fibres have been traced only for short distances 

 (Fig. 102). In a number of cases fibres have been found to pass over cross- 

 wise from one part of the cord to the other (Fig. 114). Both the radiating 

 fibres and the crossing fibres enlarge slightly toward the periphery and end 

 abruptly at the edge of the cord. 



In later stages the connection between the nerve cord and hypoderm 

 consists of spindle-shaped*, bipolar cells placed in close succession one behind 

 the other in a single row, with nowhere an indication of a ganglion (Fig. 45). 



In the male the ventral cord separates into two branches at the posterior 

 end, a branch passing into each prong of the fork (Fig. 98) and ultimately 

 disappearing in the hypoderm (Fig. 36). Beginning at about the point 

 where the connection between hypoderm and cord becomes divided and 

 passing backward to the cloacal musculature there is an enlargement of the 

 cord, the cloacal ganglion (Figs. 101, 102). It consists chiefly of an increase 

 of the fibrous part of the cord and hardly deserves the name of ganglion. 



In the female an enlargement of the cord occurs near the posterior 

 end, opposite the cloacal musculature (Fig. 92), while the cord is passing 

 into the hypoderm. This appears as an enlargement of the cellular part of 



