yOO 'Journal of Comparative Neurology and Psychology. 



According to the distribution of their subdivisions, splitting nerve 

 fibers may be of three types which we may term arbitrarily types 

 I, II and III. Let type 1 include those nerve fibers whose sub- 

 divisions after splitting follow two pathways, one subdivision passing 

 to. the branches and the other continuing in the main trunk. Let 

 type II include those nerve fibers which send both subdivisions to the 

 branches, then type III may represent the group which sends both 

 subdivisions to continue in the main trunk. Any combination of the 

 three types of splitting nerve fibers, provided the number of splitting 

 nerve fibers is not increased, would give the same apparent excess 

 of nerve fibers, namely two in each instance. The only method of vary- 

 ing the apparent excess by varying the relations of the splitting fibers 

 is by increasing or decreasing the number of nerve fibers which split. 

 If the number of splitting nerve fibers were increased the excess 

 would correspondingly increase. In the same way a decrease in the 

 number of splitting nerve fibers would lessen by that exact number 

 the excess of nerve fibers. 



If to a fixed number of nerve fibers, some splitting, some non- 

 splitting, another group of nerve fibers be added, by no combination 

 of splitting fibers in the added group could the actual number of 

 splitting fibers in the combined group be decreased. The excess 

 would be unmodified if no new splitting fibers were present in the 

 combined group, or, if splitting fibers were added, the excess would 

 be correspondingly increased. 



The excess may be modified by two conditions : the retention at 

 the level below the branches of nerve fibers which have disappeared 

 above the branches, or the presence of fibers at the higher level which 

 have disappeared at the lower level. In the first instance the excess 

 would be increased, in the second, decreased. 



Returning now to the findings for frog E, let us attempt to ana- 

 lyze the disparity between the excess for the left leg and that for the 

 right leg. Either percentage excess might be within the normal 

 variation for the frog as determined by previous enumerations, Dunn, 

 1900, 1902. It is necessary therefore to marshal the arguments for 

 and against the integrity of the nerve fibers, and hence of the 

 findings for each leg. 



