299 
cranial) ganglia does not take place, at any rate in Raja, as described 
by His in Mammals. My conclusions are also somewhat different from 
those of Dourn. The formation of the posterior columns comes into 
play as a factor in the process. There is a bulging out, cellular in 
nature, of the dorso-lateral portion ofthe cord, and this is met by 
short cell-chains reaching upwards from the ganglion. The two fuse 
together or rather become applied, and thus form a continuous chain 
of several rows of cells reaching from the ganglion for some distance 
along the route of the future posterior column !). Along this course 
the nuclei proceed to lay down fibrils in the way already described. 
e) As just seen the mode of development of the columns of the 
cord inevitably crops up. In dealing with these a portion of what 
His has termed the “Randschleier” would come under review. Ac- 
cording to him in the higher forms, though an unexplained exception 
is made on behalf of Selachians, nuclei are absent in this region during 
the development of these columns. 
My observations are not in harmony with this view. I find that 
wherever a column or tract of fibres?) arises in the nervous system, 
its development (as in the case just mentioned) is initiated by the 
laying down of chains of nerve forming cells. As already stated, in 
the case of any spinal ganglion the path from the latter along the 
posterior column may readily be proved to be laid down by the appli- 
cation of cell-chains from the ganglion to similar cell-chains within 
the nervous system. 
While speaking of the columns within the central nervous system, 
it may parenthetically be remarked that the processes in the development 
of such reveal pictures which recall in a striking fashion the things 
seen in the developing lateralis. This is true of both transverse and 
horizontal sections, and for the former what can be seen in the most 
posterior (caudal) portion of the spinal cord finds a very close parallel 
in the developing lateralis. More detailed comparison and figures 
may be reserved for another occasion. 
f) In certain stages, and these not very young ones, there may 
be observed perched on the most dorsal and outer limit of each spinal 
ganglion — in reality rather on the developing root — a little group 
of deeply-staining cells or nuclei. It is at first continuous with, and 
afterwards represents the remains of, the cell-commissure connecting any 
1) There is of course a bend where the chains from the ganglion pass 
into the bulge. 
2) Naturally I have not studied the development of anything like 
all the columns or tracts of white matter in the central nervous system. 
