THE HISTOLOGY OF DISSEMINATED SCLEROSIS. 583 



swollen axis cylinder. In the second stage the picture is very definitely one of 

 degeneration in most of the affected nerve fibres, while the few fat granule cells 

 present are filled with fine granules of a Marchi-staining substance, and numerous 

 other cells show a commencing fat granule formation in their protoplasm. In the 

 next stage the whole area is beset with large black formations — fat granule cells — 

 scattered through the tissue and grouped around blood-vessels and septa (fig. 313). 

 In the two succeeding stages the Marchi reaction is confined to the remaining fat 

 granule cells chiefly around the vessels (fig. 315), while in the final stage the 

 sclerosed tissue stains lighter than the normal, and the picture, like the Weigert one 

 at this stage, is again a negative one. 



Eeference must be made to the transitional zone of such an area ere its complete 

 evolution can be understood. The area which served as the illustration for our 

 description was a roughly triangular one, with its base to the posterior commissure 

 (cf. figs. 260 and 356). This well-marked layer of dense glia tissue around the central 

 canal seemed to act as a barrier to the further development of the area in this 

 direction, but at its point of greatest development the two lateral sides curved 

 gradually to the apex about the middle of the posterior septum. Along these lateral 

 sides there was a very gradual transition to the normal tissue which bordered the 

 posterior horns. In this transition zone the stages were very similar to those 

 described within the area, but were later in their development, so that, for example, 

 when the fibril formation had commenced within the area, there was still nothing of 

 it to be seen in the transition zone. It was, however, in the two final stages that the 

 transition zone was most marked, for here, long after no trace of fat granule cells 

 could be recognised even in Marchi preparations, there was a well-marked peripheral 

 zone of such cells, and in this zone the vessels still retain traces of fat granule cells 

 in their adventitia at a stage when the central sclerosis has been long complete. 

 The presence of this zone containing fat granule cells and other changes character- 

 istic of an earlier stage of development, e.g. degenerating nerve fibres and proliferated 

 glia cells, justifies the assumption that the process develops excentrically. When the 

 transverse section of a sclerosed area in Marchi sections shows up lighter under low 

 power than the surrounding normal tissue, and no degenerating fibres nor fat granule 

 cells can be found either in the area itself or within the vessel walls at its extreme 

 limit, the impression is given that the area has reached its climax of development, 

 and that the process is stationary. All the products of degeneration have been 

 removed, and the glia fibril formation has also reached its climax, except in the 

 transitional zone, which seems also stationary. Here a nuclear proliferation is very 

 evident, mostly of small round cells ; the nerve fibres have a thin, faintly-staining 

 myelin sheath, but not a degenerated one ; and it can be seen that the glia cell 

 enlargement is continued into the normal myelinated tissue around — the glia cells 

 showing an enlargement of their protoplasmic processes, without any apparent 

 change in myelin sheath or axis cylinder. 



