I 1 6 J. B. JOHNSTON. 



of the first, second, third and fourth nerves of normal living ani- 

 mals are to be seen rounded projections or knobs which might 

 be described as bud-like or as resembling an up-turned thumb. 

 The greatest difference between these structures and the supposed 

 spinal ganglia is that these stain in methylene-blue at the same 

 time with those bodies which are connected with the end branches 

 of the rostral nerves ; that is, some hours earlier than the sup- 

 posed spinal ganglia. 



My own observations have been made upon animals stained in 

 methylene-blue and upon sections prepared by the Golgi method 

 and by a variety of hsematoxylin methods after fixation in twenty 

 per cent, formalin, Worcester's, Zenker's or Flemming's fluid. 



In living animals stained with methylene-blue, both types of 

 cells described by Retzius and the fibers from the transverse cells 

 passing into the dorsal roots have been clearly and repeatedly 

 seen. In two specimens also, many examples of slender bipolar 

 cells were seen in the root of the nerve in the situation of the cells 

 shown in Fig. 4. Each of these cells sent one process out along 

 the nerve and one into the cord. When the cells in this position 

 are examined in hsematoxylin sections they are seen to be spindle- 

 shaped cells with elongated nuclei and with a process from each 

 end of the cell which can be followed for a considerable distance. 

 The cells have very slender bodies with only a thin layer of cyto- 

 plasm over the nucleus. Except for the difference in size, which 

 is more or less proportional to the difference in the size of the 

 two animals, these cells are closely similar to the ganglion cells 

 of the cutaneous and lateral line fibers in Petromyzon (Lampetra 

 Wilderi, 6). Such cells are found in the root and undivided trunk 

 of the nerve, in the proximal part of the dorsal and ventral rami, 

 and also in the cord near the root of each nerve. The largest 

 number of cells are found in the proximal part of the root and 

 in the somewhat conical protuberance of the cord which gives 

 rise to the root. In the so-called cranial nerves (I., II.) the large 

 size of the roots and the somewhat more regular arrangement of 

 the cells and fibers enable one to see these elements more clearly 

 than in the trunk nerves. Comparison of horizontal and trans- 

 verse sections shows that in both cranial and trunk nerves the cells 

 are distributed throughout the thickness of the nerve root. 



