544 HENRY H. DONALDSON AND G. NAGASAKA 
under ‘relative areas of body’ in table 9 (A). If, for comparison, 
we take the cubes of the mean diameters of the ganglion cells, 
which correspond to the relative volumes of these cells, and com- 
pare these in a similar manner, we obtain the series of ratios in 
the last column under the ‘relative volumes of ganglion cells.’ 
The volume ratios there given correspond very closely with the 
ratios for the increasing areas. From these relations it is fair to 
conclude that within the limits of this series the volume of the 
spinal ganglion cells increases at the same rate as the area of the 
body surface.- This result accords with the conclusion reached 
by Levi (08) and with the general results of the later work by 
Busacea (716), although the explanation of the relation here 
presented is an extension of that given by those authors. 
It is hardly necessary to repeat that the cell bodies in the 
ventral horn of the spinal cord do not enlarge in a like way, as 
can be seen by comparing the data for cell diameters in table 3 
with the corresponding data in table 6. 
In this connection another important relation appears, although 
for the discussion of it data which are not given until table 10 
must be used. 
Linking the ganglion cell body with the surface of the body, 
and other points of termination, is the peripheral fiber in which 
the axis represents the conducting tissue. It might be assumed 
that this axis would increase in its cross-section in a definite rela- 
tion to the increase in the surface to which it is distributed. 
The relations of the increasing area of the axis cylinder to the 
increasing surface of the body are shown in table 9 (B) where the 
values for the areas of the axis cylinder in the fibers just distal 
to the ganglion (at A, fig. 1) are treated in the same manner as are 
the computed areas of the body in the several groups. The 
table shows that the areas of the axis cylinder increase at about 
the same rate as the areas of the entire body. However, in the 
first instance, when the area of the axis in the smallest group is 
compared with that in the largest, the ratio, 3.71, is found to be 
less than that for the body surface, which is 4.37. It is just 
possible that the smaller increase in the areas of the axes is due 
to the splitting of the fibers in their course to the periphery, but 
{ 
