The connectives with the thoracic-abdominal ganglion were long and slender, with each its small ganglion a short distance from the brain. A cross connection between these connectives was not seen. It may have been broken in the dissection. The thoracic-abdominal ganglion has many nerves connected with it, as shown in the figure; the largest of these were traced to the legs and upper thoracic appendages. The legs are large and heavy and the nerve trunks in them are large; their combined bulk would probably be many times that of the ventral ganglion. So far as studied, the internal arrangement of tracts and cells does not differ materially from the classic descriptions of Bethe in another species. One thing especially noteworthy is the fact that the nerve cells do not seem especially large, nor are the large ones numerous. [Illustration: Figure 1] The nerve cells and fibers were studied in preparations fixed in Flemming's fluid and stained with iron hematoxylin. As in forms previously studied, the general structure of the ganglion in a way duplicates the structure of the nerve cells, in that a general reticulum forms a framework for the other structures in both. It is hard in individual cases to distinguish the supportive structures from the conductive, but the fibers and fibrils in or outside of the nerve cells run in longer straight lines--that is, they do not form so much of a meshwork, although they may branch and intertwine to some degree both within and outside the nerve cells. Large strands or fibers from nerve cells run as fibers, then divide into smaller masses of fibrils, and at last break up into numerous fibrils. The usual demonstration of nerve cells with their branches as shown by the Golgi or methylene blue methods, I believe, shows only the _larger_ and _smaller_ branches from nerve cells, and the smallest branches where the fibers break into fibrils are not shown at all. [Illustration]