ANATOMY AND CLASSIFICATION OF THE ABENICOLID^. 491 



two pairs of branches, which ruu down the sides of the cord 

 to the roots of spinal nerves. At the level of the chgetigei-ons 

 annulus the three fibres exhibit a thickening of their fibrous 

 sheaths ; they fuse laterally, and a pair of branches spring 

 from the fused mass and run to the spinal nerves of this 

 annulus (PI. 28, fig. 74). 



Behind the point of contact of the lateral giant-fibres and 

 the processes of the giant-cells the arrangement is simpler. 

 Here, as a rule, one of the lateral fibres, after a little 

 preliminary branching and crossing, disappears in the cord, 

 and the other alone continues through several annuli by the 

 side of the median fibre until the next chtetigei'ous region is 

 reached, and then the complicated arrangement of the great 

 fibres begins again. 



This is obviously a case where a topographical method is 

 required to determine with accuracy the connections and 

 terminations of these fibres, but, as we have explained in a 

 preceding section (p. 468), neither methylene blue nor Grolgi's 

 method, applied in the usual manner and varied in several 

 ways, was effective. We have, however, shown that there is 

 an intimate connection between the giant-fibres and the 

 giant-cells, between the giant-fibres and the spinal nerves, 

 and between the process of the giant-cell and the fibrous 

 matter of the cord.^ 



c. The Nature of Giant-cells and Giant-fibres. 



The histological features of these giant-cells are identical 

 with those of unipolar ganglion cells. Indeed, the only other 

 class to which the giant-cells could be referred would be the 

 large neuroglia cells of the kind described by Apathy (1897) 

 in certain leeches as " grosse mediane Sternzellen " and 

 '' Leydig's Zellen." These glia-cells, however, are not only 

 multipolar, but they exhibit throughout their cell-body a 



' We are fully aware of the opinion shared both by Retzius and Lenhossek 

 that the giant-fibres are not nervous, and do not originate in the way we 

 describe. The discussion of the evidence for and against the view we adopt 

 is given more fully on p. 492. 



