124 J. B. JOHNSTON. 



the coarser central fibers, some of them ramify at once on enter- 

 ing the cord, others divide in T-form and run in the dorsal 

 bundles of the same side, others enter the dorsal bundles with- 

 out dividing, others go to the opposite side of the cord and either 

 ramify at once or run in the dorsal bundles, with or without 

 bifurcation. Altogether, the number of coarser fibers which 

 ramify near to the root from which they come is striking. If 

 these are cutaneous fibers a striking physiological fact is explained, 

 namely the relative independence of the segments in locomo- 

 tion ; otherwise expressed, the small number of segments neces- 

 sary to perform the typical swimming movements. A short 

 piece of the tail end can swim well and behaves much as a whole 

 animal does ; and this for many days together. 



To illustrate by a complex movement, normal animals in a 

 shallow dish persistently put their heads up over the edge of the 

 dish and then by swimming round the dish and pushing against 

 the edge succeed in wriggling over, if the edge is not too high. 

 The isolated tail makes the same persistent and apparently pur- 

 poseful efforts when the dish is very nearly full of water. When 

 an animal is so macerated that all the tissues except the noto- 

 chord are gone from the middle of the body, the two parts per- 

 form typical swimming movements but each with an independent 

 rhythm. This retention of the power of coordinated move- 

 ments by a few isolated segments is perhaps connected with the 

 large number of cutaneous fibers which have a short course in 

 the spinal cord. This makes it possible for the muscles to be 

 reflexly controlled by stimuli received at the surface of the body 

 in the same or adjacent segments. 



Finally, a few coarse fibers whose ganglion cells are in the 

 nerve trunk tro to form the definite bundles described above. 



o 



Since these are chiefly ascending fibers which have a long course 

 in the cord, the bundle may be compared with the dorsal tract 

 of the same description in vertebrates, the tract of Goll. This 

 is therefore probably the first tract to appear as a definite bundle 

 in the vertebrate nervous system. 



The fourth type of cells described in this paper are the vis- 

 cero-motor cells. In position they correspond to the viscero- 

 motor column as it is known in fishes and other vertebrates. 



