570 DR JAMES W. DAWSON ON 



(figs. 9, 10), from their function of absorbing the products of the disintegration of the 

 myelin. These cells lie not only in all the tissue interstices, surrounded by the branch- 

 ing processes of the glia cells, but in the lymphatic spaces of the adventitia of the blood- 

 vessels — whither they have in all probability been drawn in from the tissue spaces 

 by the suction influence of the lymph flow in the adventitial lymphatics. Eound 

 even the smallest capillaries these cells form a complete ring, and the area under 

 low power assumes a very characteristic appearance. The presence of such large 

 numbers of fat granular cells gives to the blood-vessel walls an appearance of a cell- 

 infiltration, for in the vessels larger than the capillaries they are present in several 

 rows, often closely compressed. In addition to these cell elements in the adventitia, 

 other nucleated elements add to the nuclear abundance of the vessel wall. The endo- 

 thelium, especially of the capillaries, frequently shows evidence of a distinct prolifera- 

 tion, and in the adventitia are found dark-stained nuclei, together with nuclei of a 

 vesicular clearer character, both of which have probably arisen from the proliferation 

 of the cells of the adventitia and the endothelial cells of its lymphatic spaces. 

 Diffuse stains, e.g. Van Gieson's stain, bring out, although feebly, one further tissue 

 component — the axis cylinders. These are no longer the sharply defined, homo- 

 geneous points of the dense sclerotic tissue, but a faintly-staining, almost unrecog- 

 nisable, swollen structure, which is a lesser degree of the intense swelling of the axis 

 cylinders seen in marked oedema of the cord. These swollen axis cylinders are 

 usually found lying in the tissue meshes, with no trace, or only slight traces, of 

 myelin around them, and often closely compressed between the fat granular cells and 

 the protoplasmic processes of the large glia cells (fig. 10). The iron-hsematoxylin 

 stain shows that in this area there is as yet almost no attempt at fibril formation on 

 the part of these glia cells, although the processes at their lateral margin and 

 terminations show a definitely darker staining. 



It has been seen that Weigert sections show the almost complete absence of 

 myelin in this area ; that diffuse stains give (l) the characteristic appearance of the 

 two cell elements — proliferated glia cells with numerous processes and the fat granular 

 cells — (2) the numerous dilated blood-vessels, and (3) the persistence of numerous 

 swollen axis cylinders ; and that glia stains give as yet no definite fibril formation. 

 There remains now to be mentioned the appearance with Marchi-stained sections 

 (figs. 66, 313). This is the most characteristic of all, and has given to the areas the 

 name of " fat granule cell myelitis." Each of the large vacuolated cells in the tissue 

 spaces and walls of the blood-vessels is found to be composed of a very large number 

 of minute granules of a substance staining black with osmic acid. Most of these 

 granules are quite round, but a few show irregular contours from compression. 

 Granules are found also in the spaces between the cells. Under low power the 

 area is thickly studded with these black granular cells, which also form concentric 

 rings around the blood-vessels. No trace of normal myelinated fibres can be found 

 within the area, but in the transition zone are recognised numerous fibres in all 



