THE HISTOLOGY OF DISSEMINATED SCLEROSIS. 577 



run, numerous examples of these may be found — the large ramifying processes joining 

 the general glia reticulum. In such a condensed glia tissue the axis cylinders may 

 be recognised for a long time (fig. 345). They appear little altered either in quantity 

 or quality, but may be atrophied and gradually disappear or appear to fuse with the 

 general glia tissue. The blood-vessels present thickened, condensed walls, with few 

 cell elements. No compound granule cells can be found either distributed in the 

 tissue or in the adventitial lymph spaces of the vessels. The proliferated glia cells 

 seem to remain for a long time enlarged in this condensed tissue and their very 

 defined processes can frequently be traced for a long distance, with no relation to 

 the close feltwork of the glia in which they lie. Cross-sections of such areas are 

 seen in figs. 345, 347, in which the dense glia tissue appears almost fused and stains 

 lightly with Van Gieson's stains : the numerous glia cells are still larger than the 

 largest spider cells of the normal cord. Weigert sections of such an area show a 

 complete absence of myelin in the fully sclerotic area, but in the earlier stage of 

 thickening of the glia trabeculse, the myelin ring appears only thinned and atrophic 

 and gradually lost. A low-power view of such an area gives the impression merely 

 that the myelin sheath is faintly stained compared to the normal tissue, the transi- 

 tion to which is more abrupt than the glia-stained sections would indicate. In 

 Marchi-stained preparations of the fully developed area, there is a complete absence 

 of any indication of myelin degeneration. 



It must be noted, however, that in some areas which also give the impression of 

 a primary glia change, which has gradually encroached upon the nerve fibres and led 

 to the dissolution of the myelin, compound granular cells have been found at the 

 periphery of the areas and in the walls of the small vessels within it, and also for a 

 time very isolated examples in the tissue itself. Such cells are always very isolated 

 and give the impression that the process has been a slow one, and that somehow, in 

 the gradual withdrawal of these cells from the tissue spaces into the lymph channels, 

 a few have been left behind. 



Such areas, from the slowness of the demyelination process ; the absence of all, 

 or almost all, signs of reaction except in the neuroglia tissue ; and the frequent 

 absence of granular cells, may be looked upon as evolving gradually to a complete 

 sclerosis without any intermediate stage of fat granule cell myelitis. 



B. Evolution of an Actual Sclerotic Area through a Stage of so-called 

 "Pat Granule Cell Myelitis." 



(a) Spinal Cord. 



(i) Nerve fibres cut transversely (figs. 8-12 ; 349-354). 



It will be convenient again to take an area in the ventral third of the posterior 

 columns of the cervical cord, for such are very numerous in our preparations. To 

 trace the gradual changes by which the normal tissue is replaced by sclerotic tissue 



