580 NEKVOUS SYSTEM. 



reagents, which must produce more or less change in the structures ; and it seems an 

 anatomical impossibility to make these sections so as to follow out the prolongations of 

 the cells far enough to establish beyond doubt their exact relations. 



These two considerations alone are sufficient to account for the uncertainty so appar- 

 ent even in the most successful investigations into the anatomy of the central nervous 

 system ; and we shall content ourselves, in view of these facts, with giving a summary 

 of what seems to be the probable relation of the cells to the fibres of origin of the nerves 

 and to each other. 



Apolar cells, if they exist at all and be not cells from which the poles have become 

 separated, are simple, rounded bodies, lying between the fibres, with which they have no 

 other relation than that of mere contiguity. Unipolar cells have but one prolongation, 

 which is continuous with a nerve-fibre. It is not certain that these exist in the human 

 subject. 



Bipolar cells are found in the ganglia of the posterior roots of the spinal nerves and in 

 some of the sympathetic ganglia. In many of the lower animals, particularly in fishes, 

 the cells of the ganglia of the spinal nerves are simple, nucleated enlargements in the 

 course of the sensitive nerve-fibres, and many anatomists have inferred that the same 

 arrangement exists in man and in the mammalia ; but the constitution of these ganglia in 

 the higher classes of animals seems to be entirely different. In the first place, the roots 

 of the spinal nerves at the ganglia are undoubtedly reenforced by the addition of new 

 fibres, as Kolliker has shown by actual measurement, the roots being sensibly larger 

 beyond the ganglia, while the filaments of entrance and exit have the same diameter. 

 Direct observation upon the ganglia in man also fails to show the arrangement which is 

 so clearly demonstrable in fishes. The cells in the posterior roots are not continuous with 

 the fibres passing from the periphery to the cord, but they give origin to new fibres, 

 generally two in number, which sometimes are single, and sometimes bifurcated, and 

 which pass, in by far the greatest number of instances, if not in all, to the periphery. 



The inultipolar cells, with three or more prolongations, are found in all of the ganglia, 

 but they predominate largely in the gray matter of the cerebro-spinal centres. It is the 

 question of the exact mode of connection between these cells and the fibres of origin of 

 the cerebro-spinal nerves and the union of the cells with each other by commissural pro- 

 longations, that presents the greatest difficulty and uncertainty. One point, which has 

 been raised within a few years, is with regard to the character of the different poles 

 connected with the same cell. In ordinary preparations of the central nervous system, 

 it is impossible, even with the highest available magnifying powers, to distinguish any 

 one pole which, in its general characters and connections, is different from the others ; 

 yet, some anatomists describe a single pole, more distinct in its outlines than the 

 others, which does not branch and is to be regarded as an axis-cylinder. The other poles 

 are supposed to be of a different character, not connected with the nerve-fibres, and 

 always presenting a greater or less number of branches. These views are accepted by 

 Schultze, who gives a figure, after Deiters, in which the contrast between the poles is 

 represented as very marked ; but, although this opinion is accepted by other high authori- 

 ties, it is not easy to understand how it can be received without reserve, when it is so 

 difficult, if not impossible, to follow out the poles, except for a very short distance. 



With our present means of investigation, there seems to be no doubt with regard to 

 the following facts : Tracing the nerve-fibres toward their origin, they are seen to lose 

 their investing membrane as soon as they pass into the white portion of the centres, being 

 here composed only of the medullary substance surrounding the axis-cylinder. They 

 then penetrate the gray substance, in the form of axis-cylinders, losing here the medul- 

 lary substance. In the gray substance, it is impossible to make out all of their rela- 

 tions distinctly, and we cannot assume, as a matter of positive demonstration, that all 

 of them are connected with the poles of the nerve-cells. Still, it has been shown, 

 in the gray matter of the spinal cord, that many of the fibres are actual prolongations 



