42 EDGAR J. ALLEN. 
Two other fibres in the same region often stain, and are 
shown in fig. 4, fandg. The fibre f appears to enter by the 
posterior nerve-root of Th. 111, to pass transversely across the 
ganglion, and leave by the posterior root of the opposite side, 
the straight course being broken by a slight indentation at the 
middle line. It is not unlikely that there are in reality two 
elements which decussate in the middle line, and are there 
connected with cells which have remained unstained. The 
fibre g has a similar relation to the posterior roots of Th. rv. 
Traces of a system of elements similar to that just described 
occur in Th. 111. 
A pair of elements exists in each ganglion from Th. vi to 
Th. x, which unfortunately seldom stains, but when stained 
presents several points of interest. The appearance is generally 
that shown in fig. 1, Th. rx, W., but in one preparation the 
element stained on one side only of the ganglion, and fortu- 
nately showed the position of the cell. This element, which 
actually occurred in Th. vr is inserted in fig. 1 for the sake of 
clearness in Th. x. (W.). Since in another preparation in 
which the element was stained, a cell just commenced to take 
up a similar position, I feel little doubt that this figure (fig. 1, 
Th. x., W.) represents the true condition of the element. Its 
course may then he described as follows :—Starting from a 
cell, which lies in the anterior portion of the central mass of 
ganglion cells, the fibre passes first backwards, decussates with 
its fellow of the opposite side, and then turns outwards to the 
neuropile, upon entering which it gives off a small tuft of fine 
branches. After giving off this tuft, the fibre continues to run 
transversely through the neuropile to about the level of the 
lateral mass of ganglion cells, where it bends suddenly forwards, 
giving off a short richly arborescent branch at the angle, and 
then runs to the next ganglion in front, in the neuropile of 
which it ends in another tuft of fine branches. No branch 
has stained which passes out through either of the nerve-roots. 
If the above description represents in reality the entire course 
of the element, it must serve to put different parts of two 
adjacent ganglia into communication with each other. 
