THE HISTOLOGY OF DISSEMINATED SCLEROSIS. 643 



gone an excentric spread comparable to the central gliosis. The end result, then, 

 of all such areas would probably have been a compact gliosis in which were very 

 fine meshes not in the least similar to those of the areolar areas described by 

 Muller as characteristic of secondary disseminated sclerosis. The examination of 

 many hundreds of areas has satisfied me that "early" areas— so-called areas of fat 

 granule cell myelitis — can develop into areas of compact gliosis with all the 

 characteristics of those described by Muller as typical of true, primary disseminated 

 sclerosis. None of the cases could be described as "acute multiple sclerosis," taking 

 this term in the sense Marburg and Dinkler have used it. In Case I, which has 

 the shortest clinical course in the series, the illness had lasted fifteen months and 

 death had resulted from diarrhoea and exhaustion. In this case no fewer than fifty 

 different areas were examined by the Marchi method alone, and four-fifths of these 

 showed fat granule cells distributed throughout the whole affected tissue. In the 

 other areas these cells were limited to the periphery and to the vessel walls, while 

 the centre of the area was composed of a dense glia feltwork with numerous axis 

 cylinders in the fine meshes. In only two areas was there a complete absence of 

 any such signs of inflammatory reaction. In the very numerous areas examined by 

 other staining methods, a similar structure could be demonstrated and in similar 

 proportions, for the areas chosen for the Marchi method were taken irrespective of 

 their macroscopic soft or hard consistence. 



I am in agreement with Muller's statement that the glia tissue had frequently 

 not the same abrupt definition as the demyelinated tissue. Though not using 

 Weigert's glia stain, it was often quite evident that the myelin sheaths at the 

 margins of the area, in longitudinal sections, were thrust asunder by proliferated 

 glia tissue. This, however, need not necessarily be explained on the ground of a 

 primary glia proliferation (see p. 666). I am further in agreement with Muller 

 that the most frequent sites are the peri-ventricular and peri-central tissue, around 

 the postero-median septum, in the lateral columns, and with the marginal glia zone 

 as a base. Why certain parts are predisposed it is difficult to say, and this question 

 must be discussed later. More easy to understand is the frequent remarkable 

 symmetry emphasised by Muller. This has been entirely confirmed, especially 

 in cases where there was an almost complete transection of the cord, when the 

 symmetry was perceptible in the marginal portions of preserved fibres, and again 

 when both the lateral columns were affected and marginal zones were left corre- 

 sponding to the dorsal cerebellar tracts. Muller's explanation of this symmetry 

 has been referred to, but it may also be explained by the possibility of the two 

 halves of the cord being exposed at the same time to a diffusely acting agent in 

 the blood-vessels. 



Muller's essential arguments are (l) that the participation of the blood-vessels 

 within the area is only a secondary one, and (2) that the glia proliferation is far 

 more than reparatory. In discussing the origin of the process we must refer to 



