622 DR JAMES W. DAWSON ON 



from the adjoining white matter in radiating lines into each medullary ray. In 

 the transition zone between cortex and white matter the affected area often 

 assumed the form of an elongated spindle, and when this zone lay in the cup of 

 a convolution the spindle curved on itself, with the concavity to the fissure. In 

 the extension of either flat or curved spindle, the poles often passed to involve 

 the transition zone between two or more adjoining convolutions, and in this 

 extension the whole of the white matter from which the individual medullary 

 rays arose was often cut across. 



In the large basal ganglia, isolated areas were often quite round (fig. 266), but 

 the fusion of primary areas in the optic thalamus, internal capsule, and lenticular 

 nucleus produced large irregular areas in which might be traced the original forms. 

 Symmetrical involvement of the basal ganglia and especially of the external capsule 

 and claustrum, with an extension to the grey matter of the convolutions of the island 

 of Reil, formed a prominent feature in several of the sections, especially in Case II. 

 The distribution of the areas in the basal ganglia in individual cases has been given 

 elsewhere, but here it may be noted that such areas seemed often to start in a peri- 

 vascular zone around the lenticulo-striate and strio-thalamic vessels (fig. 267), and 

 that the peri-vascular sieve-like areas were nowhere so marked as in this region, 

 around the larger of these branches. 



In the cerebral cortex the variation in the form of the areas is even more marked 

 than elsewhere. These may be grouped into three divisions according to whether 

 the demyelination (l) spreads from white matter into grey and is arrested within 

 the radiations of the nerve fibres, or (2) is wholly within the cortex with a zone of 

 intact nerve fibres forming a fine network superficial to it, or (3) spreads from the 

 surface and extends for a variable distance inwards. The shape of those extending 

 from the white matter is usually dependent upon the subcortical portion, and the 

 extension may be simply a gradual one which involves the radiations in a sinuous 

 or curved or even pointed outline. The definition of the cortical portion within the 

 radial fibres is often very sharp, even under a high power, but frequently individual 

 fibres or even the groups of fibres forming a single radiation may pass in from the 

 white matter into the area for a considerable distance. In the areas, situated as far 

 as could be made out by numerous serial sections, entirely within the cortex, no 

 special shape could be noted. It was possible to trace small areas entirely within 

 the tangential fibres (fig. 300), or involving the fine fibres forming the supraradial 

 network or cutting across the Baillarger's or Genari stripe, but more often these 

 small isolated areas were found limited to the intra-radial and radiating fibres, 

 which pass, with a fan-like arrangement, from the tips of the medullary rays. 

 In many such areas a small vessel, which seemed to be the centre of the demye- 

 lin.ition, was certainly found, but others bore no relation to any of the numerous 

 capillary vessels found within them. The shape of the areas, passing in from 

 the surface of the convolutions, was often that of a modified wedge, or of an 



