502 
David H, Dolley 
the first abdoiiünal gaiiglion* there were oiily two series froni this sensory 
type, aud as tliey do not include all sack cells, it is possiblc that no multi- 
polar cells were measured, even if present. 
In addition to tliese foiu‘ main groiips, there is a fifth order which 
must be differentiated. These are the smallest ceUs of all. A series of 
seventy-five was measured, though only for two diameters, and they 
have a nucleus-plasma coefficient much smallcr than that common to 
the resting types of the main groups. This in itscif puts theni in a dif- 
ferent dass, but in addition to their size and nucleus-plasma relatioji, 
they are further distinguished by having in addition to the karyosome 
a moderate number of basichromatin granules in the nucleus, located 
at the nodal points of a more distinct nuclear reticulum and more nu- 
nierously at its periphery (Fig. 23). The other types have no basichromatin 
outside of the karyosome. These cells are regai'ded as probably associa- 
tion cells. Together with the other cells they show the same effects of 
activity and the measurements give sufficient indication that the trend 
of their size changes is substantially the same. 
Only in the supra-oesophageal ganglion does a sixth order of cells 
appear to be superadded. In these the nuclei are most in evidence and 
they are arranged in discrete groups. They appear to correspond to 
the cytochrome type of Nissl (1895). 
It would seein more likely that the multipolar cells referred to 
belong in this category of association cells. The writer is more 
influenced toward this opinion by the nature of the cellular rcactioii, 
apart from a consideration of the number and distribution of what 
may be termed primal sensory and motor functioning cells. From the 
anatomical unity of the process which results in all types hi response 
to Stimulation, there is no need for additional complexity of primal types 
to conserve varieties of either Sensation or motion. This idea is further 
l)orne out by the absence of niultipolar cells in the more simple first 
abdominal ganglion as stated by Retzius. Here the fifth order just 
described suffices for the connecting links. It would then be expected 
that with greater complexity, a differentiation of associational neurons 
lias arisen, characterized by structural differences which are esscntially 
those of niultiplicity of Connections. The reasoning upon which this 
conception is based will be more appropriately elaborated in the section 
on the mechanics of nerve cell activity. 
A most interesting point worthy of incidental note is that the number 
of cells of groups and sub-groups would appear to be a constant. In 
measuring the motor cells from the first abdominal ganglia of five animals. 
