Here the cell substance, as Max Schultze remarks, “is a continuation of the axis cylinder, and encloses the nucleus. The medulla commonly ceases at the point where the axis enters the cell, to reappear at its exit; but it sometimes stretches across the cell to enclose it also: so that such a ganglion cell is in truth simply the nucleated portion of the cylinder axis.”[165] There are many places in which fibres are thus found with cells inserted in their course as swellings: in the spinal ganglia of fishes these are called bipolar cells; they are sometimes met with even in the cerebellum; but oftener in peripheral nerves, where they are mostly small masses of granular neuroplasm from which usually a branching of the fibre takes place. The point to which attention is called is that in some cases, if not in all, the nerve-fibre is structurally continuous with the cell contents. The two organites--fibre and cell--differ only as regards the nucleus and pigment. Haeckel, who affirms that in the crayfish (_Astacus fluviatilis_) he never saw a cell which did not continue as a fibre, thinks there is always a marked separation of the granular substance from its “hyaline protoplasm,” and that only this latter forms the axis cylinder. But although my observations agree with this as a general fact, I have seen even in crayfish the granular substance prolonged into the axis cylinder; and in other animals the granular substance is frequently discernible. Indeed it may be said that anatomists are now tolerably unanimous as to the axis cylinder being identical with the protoplasmic cell substance. If this be so, we have only to recall the principle of identity of property accompanying identity of structure, to conclude that _whatever properties we assign to the cells_ (unless we restrict these to the nucleus and pigment) _we must assign to the axis cylinders_. We can therefore no longer entertain the hypothesis of the cells being the fountains or reservoirs of Neurility; the less so when we reflect that cells do not form the hundredth part of nerve-tissue: for even the gray substance bears but a small proportion to the white; and of the gray substance, Henle estimates that one half is fibrous, the rest is partly cellular, partly amorphous. Those who derive Neurility from the cells, forget that although the organism begins as a cell, and for some weeks consists mainly of cells, yet from this time onwards there is an ever-increasing preponderance of cell-derivatives--fibres, tubes, and amorphous substance--and corresponding with this is the ever-increasing power and complexity of the organism.