THE HTSTOLOGY OF DISSEMINATED SCLEROSIS. 641 



factors which seem to produce in any other organ the development of a process com- 

 parable in any way to disseminated sclerosis as a strong argument in favour of its 

 endogenous or developmental origin. By tracing true disseminated sclerosis to 

 " abnormal congenital conditions," he simply differentiates it from forms due to 

 recognised exogenous factors. He is in no way satisfied with this explanation, and 

 states that further investigation must seek to discover the organic basis of such 

 abnormal congenital conditions. 



It is thus seen that the explanation of a "multiple gliosis" in Strumpell's and 

 Muller's sense is very similar to that given for certain groups of tumours, which must 

 be related to embryonal defects. Eitchie, discussing the etiology of tumours, states 

 that the soundest ground for assigning congenital defect exists when the tumour 

 arises in some part of the body where at some stage of fcetal life the cells of one 

 tissue must push aside those derived from another in order to attain their ultimate 

 natural position : the sequence being that during the building up of the body the 

 cells thus detached or pushed aside, after being embedded for a longer or shorter 

 period, take on a vegetative activity and assume the characters of a tumour. 

 Muller claims that the favourite sites for the development of this gliosis are those 

 which in the embryonal period show specially active processes of growth, or where 

 two surfaces come together and fuse in later development (" Kielstreifen "), or where 

 the marginal zone of the embryonic nervous system is pushed inwards by the 

 ingrowing vessels. At such sites developmental disorders, such as germ invagina- 

 tion or detachment, might set in very readily. The ventricular surfaces and the 

 peri-central tissue normally contain much glia, the postero-median septum and 

 the optic chiasma are formed by such " Kielstreifen," and the marginal glia zone 

 and the peri-vascular glia zone represent tissue carried inwards during development. 

 Areas are, therefore, not necessarily related to blood-vessels, and when so related 

 it is in virtue of the peri-vascular glia layer which surrounds their adventitial sheath. 

 The frequent symmetry of the areas is related similarly to the glia and not to the 

 distribution of a toxin by the blood-vessels. 



The sequence of the process, as has already been pointed out, is a gradually 

 increasing hyperplasia of the abnormal glia in these sites of predilection. This 

 results in a disappearance of the myelin sheath of the nerve fibres, partly through 

 direct compression, and perhaps indirectly by derangement of the circulation. No 

 fat granule cells need necessarily appear at first* but in the zone surrounding the 

 compact area there may be a secondary reaction to the products of degeneration, 

 leading to a combination of primary and secondary proliferation with peripheral 

 progress of the process. Here fat granule cells would appear as indications of a more 

 recent process but by no means of an exogenous inflammatory one. The blood- 

 vessels within the sclerosed area also show changes which are entirely secondary to 

 the sclerotic process, which, in consequence of the peri-vascular glia layer, is 

 specially marked immediately around the vessel. 



