752 DR ALEXANDER BRUCE AND DR JAMES W. DAWSON ON 
The further structure of the nodules will be indicated in studying their mode of 
formation. 
Mode of Formation.—In order to study the mode of formation, serial sections of very 
numerous nodules were investigated. The Weigert preparations were the most useful 
for this purpose, but silver and Van Gieson sections gave very valuable confirmatory 
results ; the latter being specially helpful in indicating the mode of termination of the 
fibres. The method adopted was to trace such a nodule as that represented in fig. 35, 
stained with Weigert’s method, or fig. 29, stained with silver, both upwards and 
downwards as long as any trace of it could be noted. The first nodules investigated 
were those in the above figures, and it will be convenient to take them as representa- 
tive of a very large number of nodules which could with certainty be confirmed to 
have a similar origin and extension. 
The large wedge-shaped and oval nodules represented in fig. 35, lane in the pia 
opposite the ligamentum denticulatum, and extending inwards from the periphery, 
when traced downwards by means of Weigert preparations, were found to change 
rapidly. The fibres diminished in the peripheral mass, and the nodule increased in 
size in the lateral vessel, and, later, was connected with the pia by means of only a 
few fibres. Lower still these were absent, and the nodule, still in relation to the 
vessel but diminishing in volume, was found on the border of erey and white matter 
(fig. 36). In the next sections it was independent of the vessel (fig. 37), and its 
fibres gradually unweaving, as it were, from the nodule mass, became lost in the mesh- 
work of fibres of the grey matter; no definite connection between them could be 
traced. In the final section, only two or three strands were present, forming a very 
loose network of interlacing fibres, which in the next section could not be distinguished 
from the normal fibres of the part. Throughout its course from the periphery the 
fibres were found intertwined, and there were attempts at whorl-formation with strands 
of six to twelve parallel fibres. 
If this series is traced upwards, it is found that the nodules which lie partly 
parallel to the periphery of the cord and partly at right angles (tig. 35) are wholly 
within the pia—forming one long elongated nodule in the centre, of which at one end 
a vessel is cut transversely. This elongated nodule shows many parallel strands and 
intertwining fibres, and can be traced in higher sections nearer and nearer to the 
anterior roots, the fibres becoming more parallel (fig. 39) and slightly more deeply 
stained as the emerging anterior root zone is reached. Here the appearance of the 
bundle of fibres conveys the impression that an anterior root bundle, instead of passing 
directly outwards, has curved round into the pia laterally. Going back on the former 
figures, we see that it has travelled either in the pial spaces or in the vessel-walls till 
it had been arrested opposite the ligamentum denticulatum, that it there passes 
inwards along the lateral vessel, till it reached the zone bordering the grey matter, 
into the general texture of which it had finally unwound its fibres. The successive 
stages of this evolution may be followed in figs. 35-39. 
P bys 
arses net bes 
