THE HISTOLOGY OF DISSEMINATED SCLEROSIS. 663 



beri-beri, and other conditions, pick out the posterior columns, or both posterior 

 columns and the posterior part of the lateral columns. In pernicious anaemia, at 

 least, Minnich and Nonne have traced such degenerations to primary focal areas 

 in the posterior or lateral columns : such areas, later, either from the cumulative 

 effect of several foci at different levels in relation to the same strands of fibres, or 

 from the later involvement of the axis cylinders, are succeeded by secondary 

 degeneration. In such conditions the toxin postulated must choose out certain 

 definite vessel territories, and the symmetry is explained by the two halves of the 

 cord being exposed to the same diffusely acting agent. Numerous writers, however, 

 suggest that prolonged toxaemic states lead to primary degeneration of those parts 

 of the cord which are apparently more sensitive to trophic disturbances, and either 

 that the posterior and lateral columns have a poorer vascular supply or that, as the 

 parts most affected are the neuraxons of the posterior spinal ganglia and those of 

 the cortical motor cells, therefore the long terminal filaments of the cervical and 

 dorsal cord, they are far removed from the influence of their trophic cells. When 

 it is urged that certain areas of the nervous system may have a " general non- 

 resisting power," it can be understood how such functional factors as overstrain 

 may affect definite strands and areas, related to definite functions, but this view, 

 however unsatisfactory and unconvincing even for such conditions as cause com- 

 bined column degeneration, carries no weight when related to areas with the hap- 

 hazard distribution found in disseminated sclerosis, for one of the characteristics 

 of the disease is that the areas bear no relation to any functioning tract or 

 group of cells. 



The sequence of events in disseminated sclerosis is probably that in the 

 circulating blood the "noxa" escapes from the capillaries and transition vessels, 

 and acts upon the tissues in the sense that the area of supply of the vessel is 

 affected, i.e. it passes over into the tissues with the nutritive fluid. Numerous 

 recent experiments on man and animals tend to prove that, in general, drugs 

 administered by the mouth or subcutaneously do not pass into the cerebro-spinal 

 fluid, and Mott has pointed out the significance of Lewandowsky's observations, 

 which show that very much smaller quantities of these same drugs and bacterial 

 toxins injected into the cerebro-spinal fluid of the sub-arachnoid space produce 

 much more marked and a more rapid onset of symptoms. These and other 

 observations are taken to prove that toxic substances are unable to pass from the 

 capillaries into the lymph spaces of the nervous tissue. Goldmann's investigations 

 on the central nervous system by vital staining show that vital stains, if introduced 

 by means of subcutaneous or intravenous injection, are kept back by the choroid 

 plexus : that from the plexus the cerebro-spinal fluid receives important metabolic 

 products which are carried to the nervous tissues by the fluid ; and that the plexus 

 possesses the power of protecting the fluid, and in this way the nervous substance, 

 from the penetration of toxic substances. The role of toxins in the production of 



