ORGANS OF THE BODY. 579 



ganglion cells (d) are ooserved, imbedded in the sustentacular substance 

 of the part. These are not unfrequently tinged with brown pigment, 

 and vary considerably both in shape and in the number of their pro- 

 cesses. They are specially numerous at the apex of the anterior cornu, 

 where they usually form several dense clusters as it were. Here they are 

 separated from one another by interposed broad nerve fibres. Other 

 scattered multipolar cells, however, are met with singly, and especially 

 towards the surface of the grey substance. In the most internal portions 

 of the cord also, near the axis, as also at the base of the posterior cornu, 

 we find them still presenting precisely the same essential characteristics, 

 though decreased in size. . 



The numerous processes of these ganglion corpuscles spread them- 

 selves out in all directions, and, as a rule, are soon lost to view by dipping 

 into other planes. As observed by Deiters, whose statements we here 

 follow in regard to many points, these processes may penetrate into the 

 radiating septa of connective-tissue running through the white substance; 

 others, also, may be regularly looped around bundles of nerve fibres in 

 certain cases (Clarke, Deiters). 



These groups of multipolar ganglion cells have been very commonly 

 described as connected with one another by means of some of their rami- 

 fications, and great stress has been laid upon the importance of the latter 

 as commissures. It cannot be denied, however, that a deplorable misuse 

 has been made of this supposed existence of connecting fibres (fig. 305, p. 

 314). and it is only extremely rarely that a perfectly unmistakable view can 

 be obtained of them. Thus, in the works of many authors it is openly con- 

 fessed that, with all their efforts, they were never successful in obtaining a 

 sight of anything of the kind (Goll, Koelliker). Others even deny the 

 existence of such commissures altogether (Deiters). Others, again, are 

 able to state that they have observed them, but in rare instances (Reissner). 

 Our own experience coincides with that of the latter. Even Dean, a very 

 sound observer, who is, notwithstanding, somewhat too profuse with such 

 commissural processes, only speaks of them as exceptions. 



A second widely received axiom in the anatomy of the spinal cord, is 

 that other processes of the ganglion cells become the axis cylinders of the 

 nerve fibres of the anterior roots. This also is asserted on many sides 

 with great certaintj 1 - to be quite easy to see, whereas it is a matter of the 

 greatest difficulty in reality to obtain even one clear instance of it, some 

 observers honestly confessing their ill fortune in this respect (Goll). As 

 a rule, under favourable conditions, one such process may perhaps be 

 observed uniting with one of the motor root-bundles (Clarke, Dean, Ger- 

 lach, Frey}. 



Deiters, a recent and very thorough investigator, has added much to 

 our stock of knowledge on this very abstruse subject of the relations of 

 the central ganglion cells. We have already referred to his important 

 and repeatedly confirmed discovery ( 179), that the ramifications of the 

 ganglionic bodies are of two kinds (fig. 558). In the first place, wide 

 branching processes of protoplasm (b) present themselves, and then for 

 every cell another smooth undivided one (a), the axis cylinder process. 

 This observer, nevertheless, states that only in exceptional cases was he able 

 to follow up the latter for any distance in a section of the spinal cord. 



As may be seen in our plate, there also spring (usually at right angles) 

 from the broad protoplasm processes of the cell, a number of other very 

 delicate fibrillae. These Deiters regards as a system of secondary axis 



