The ganglion of the pro-thorax, which is greater in width than it is in length, bears a vague resemblance to a cone the base of which is turned towards the sub-œsophageal ganglion, whilst the apex points towards the second ganglion of the thorax. From the lower part spring two large nerves, their starting-point being nearer the ventral than the dorsal surface, a fact clearly comprehended when we find that the fibres of these nerves extend for the most part into the first pair of feet, that is to say, into those organs that lie underneath the horizontal plane of the ganglion. The connective filaments which penetrate the ganglion anteriorly enter it nearer the dorsal surface than the ventral, this last being extremely convex. Dissection throws no additional light upon the anatomy of the ganglion. But by means of a series of sections, we find that it is composed of a mass of fibrillar substance which occupies its centre portion and of a layer of nerve-cells surrounding the fibrillar substance. This fibrillar mass is, owing to its great volume, far the most important, and constitutes in itself alone about four-fifths of the organ. The fibrillary structure can only be satisfactorily analysed by using on it osmic acid, or other equivalent reagents which dissociate it and admit of its being reduced to a certain number of clearly differentiated elements. Whenever osmic acid or a similar reagent has not been employed, or has not sufficiently penetrated the ganglion, owing to the obstacle presented by a thick conjunctival covering or envelope, the fibrillar substance takes on a homogeneous aspect that effectually renders all analysis of it impossible. Everything depends on the employment of a good method of preparation. When the ganglion has been properly prepared, we perceive a very material difference in the appearance of the fibrillar substance when we compare the dorsal with the ventral region of the ganglion. We can do this very satisfactorily by a longitudinal section, extending through both regions. In such a section close to the median line but not confounded with it (see Cut 16)[9] we perceive that the ventral region is occupied by a cord or string of substance which owing to the action of the osmic acid has become very black, and which is formed of so dense a tissue, that we can with difficulty separate it into fibres and fibrillæ. This cord, which, by reason of its position and shape, I propose naming the _ventral column_, extends over the ventral surface of the ganglion in a longitudinal direction; at both its anterior and posterior extremities it is carried on by fibres extending into the ventral columns of the other ganglia, in such a manner that the entire series of ganglia are united by one continuous ventral cord.