608 DR JAMES W. DAWSON ON 



many ganglion cells at very numerous levels retained an almost normal nuclear 

 structure and an almost complete retention of the chromatophile granules (fig. 410). 

 The only recognisable change was an absence of the processes, a rounding of the 

 cell body, and an increase in the cell pigment. At other levels, all the cells present 

 showed a marked diminution in volume (fig. 409). and a marked pigmentation 

 (fig. 413), including cells which had become already transformed into a rounded, 

 small, non-nucleated mass of pigment. These changes occur in areas in which the 

 sclerosis is not very advanced and the intercellular tissue is composed of a net- 

 work of capillaries and branching glia cells. In a later stage such cells become 

 wholly lost in the sclerotic tissue, or small traces of pigment may still be found. 



Another type of change, and a more frequent one, is a gradually advancing 

 atrophy (figs. 415, 416), proportionate to the increasing density of the sclerosis. 

 This first affects the minute structure of the processes, and then reaches the cell 

 body, in which the chromatophile granules become powdery or are dissolved, and the 

 nucleus becomes peripheral. Such atrophying cells may assume very varied shapes 

 according to the intensity of the process, and traces of non-nucleated cells with com- 

 plete chromatolysis may be found in the sclerotic tissue for a long time. In the cells 

 of Clark's column, where we have normally an excentric nucleus and a peripheric 

 disposition of the chromatophile granules, the nucleus undergoes a marked shrivelling 

 and condensation, or at times a vacuolation before its extrusion. The cells very fre- 

 quently assume a spindle shape, and have been described by Nissl as " Fisch " cells. 



The mechanism of these changes in the spinal cord cells seems to be a simple 

 atrophy, and one factor in its causation is probably compression on the part of the 

 developing cells and fibres of the glia (fig. 419). Nowhere was there found any evi- 

 dence of the accumulation of small round cells around ganglion cells in the cord, such as 

 are to be described later around the ganglion cells of the cortex. The changes described 

 are those related to a slow, chronic process, but more extensive and general changes, 

 e.g. complete chromatolysis (fig. 411) or deeply-staining cytoplasm, are also found, 

 which must be related to the general somatic disturbances. The " coagulation 

 necrosis," described by Marinesco as an acute cellular change in which there is 

 a dense fusing together and coagulation of the individual constituent elements, 

 is rarely present. 



The cells in the different cranial nuclei undergo changes which are closely 

 analogous to those described above. The cells of the hypoglossal nuclei are fre- 

 quently very pigmented, and have the pigment distributed throughout the whole 

 cell. The glia cell proliferation may be very marked before there is any appreciable 

 atrophy of the cells, but as the sclerosis advances, here, too, there is a gradual 

 diminution in their number (fig. 418). Cells with normally staining granules may 

 be found alongside cell remnants or pigment accumulations. 



The cells in the cortex, in a demyelinated area, when involved, are much more 

 uniformly so than those of the cord. The large pyramidal cells are never normal 



