THE LUMBAR ENLARGEMENT OF THE SPINAL CORD. 345 
with cells just at the margin of the cornu, but usually enter deeply into the gray sub- 
stance, continuing, as will be shown later, in a more or less transverse course. "These 
fibres appear to be much finer than those of the other classes, and it is therefore diffi- 
cult to estimate their numerical proportion ; they are certainly less numerous than those 
of the first class. 
The fibres of the third class, or oblique descending, are quite numerous (Fig. 8, 
c, c); they sometimes enter the cord in company with an ascending bundle, following 
this for a short distance and then bending round, often at a very sharp angle, pursuing 
an oblique descending course, the exact counterpart of that followed by the oblique 
ascending bundles. Sometimes the bundle will be seen to divide (Fig. 8, c ) part of 
the fibres immediately entering the gray substance, while a portion plunge farther 
downwards, thus entering the gray substance at different depths below the starting- 
point. That all these fibres finally enter the gray substance hardly admits of doubt, 
since no one could suppose that fibres were indefinitely prolonged downwards, what- 
ever theory might be assumed regarding the ascending bundles. This is fully in 
accordance with the experiments of Brown-Sequard,* showing that all the descending 
fibres enter the gray substance within a distance of about five centimeters from their 
starting-point. I cannot, however, agree with Brown-Sequard that these are more 
numerous than the ascending fibres; it is sometimes difficult to say with certainty 
whether a bundle ascends or descends, but notwithstanding this uncertainty the ascend- 
ing fibres seem considerably more numerous. t 
The fourth class, or looped recurrent fibres (Fig. 8, d, d), seem to unite both ascend- 
ing and descending bundles. Their course is usually as follows: starting from the 
ascending end of the bundle, the fibres penetrate the posterior column as far as the 
margin of the gray substance; here the bundle often divides, part of its fibres passing 
upwards, whilst a part penetrate the gray substance, pursuing a slightly oblique, 
ascending course; they frequently proceed upwards for a considerable distance, finally 
looping round and re-entering the white column, joining some bundle with which they 
pass out, either the root immediately above their entrance, or, as is often the case, one 
higher up. In tracing the course of such a looped bundle, it is impossible to say 
* Comptes Rendus de la Soc. de Biologie, 1855, p. 79. 
T Within the gray substance most of the fibres from the posterior roots, after traversing the substantia gela: ` 
tinosa, pursue a descending longitudinal course for some distance, a will be seen when we consider the central 
course of the nerve-roots, and this may possibly be the explanation of Brown-Sequard’s experiments and 
deductions. 
