346 THE LUMBAR ENLARGEMENT OF THE SPINAL CORD. 
whether it is ascending or descending, origin and exit being obviously merely descrip- 
tive terms when applied to these fibres. These recurrent bundles have already been 
mentioned by Clarke,* in describing the oblique fibres. “Many of them,” he says, 
“both singly and in small bundles, may be observed to form loops by returning to the 
white columns.” Clarke also mentions, what I have often observed, that a few of the 
fibres from the oblique bundles * proceed near the surface both upwards and down- 
wards, and pass out again with the roots above and below them." These are also 
plainly seen in transverse or obliquely transverse sections. Stilling denies the exist- 
ence of looped returning fibres, assuming that Clarke had been deceived by confound- 
ing ascending with descending fibres (loc. ci, p. 1186); this, however, is far from 
being the case. I have been able repeatedly to verify Clarke's statement, and have 
frequently been able to trace such bundles or fibres from root to root, both in longi- 
tudinal and transverse sections. These looped fibres appear to occur frequently in the 
nervous system of some of the lower animals, especially in Lumbricus terrestris, as 
described by Clarkej and Faivre,t in the abdominal cord of which they are very 
numerous. _ 
The above description of the fibres derived from the posterior roots agrees mainly 
with Stilling’s, differing, however, in regard to the fourth class. My reason for not 
following the classification of Clarke has been, that I felt obliged to arrange the 
fibres entirely with reference to their course within the posterior white columns, from 
the conviction that neither of the three central courses according to which Clarke has 
mainly classified them originated from any one particular set of fibres belonging to 
the posterior columns. 
The Central Course of the Nerve-Roots. — In longitudinal sections made parallel to 
the median fissure, the substantia gelatinosa is seen to be traversed by numerous bun- 
dles of fine fibres, running transversely through it, nearly parallel to each other. 
The three different classes of fibres mentioned above — viz. ascending, descending, 
and transverse — pass indiscriminately into these central transverse bundles (Fig. 8), 
which are, as Clarke has already noticed, often somewhat fusiform in appearance. I 
have uniformly found that these bundles pursue the same central course in the lumbar 
enlargement which he has described with so much accuracy in the cervical enlarge- 
ment of the cat ;§ after traversing the substantia gelatinosa in a direction nearly trans- 
verse, they curve round, some upwards, some downwards, running a considerable 
distance as longitudinal fibres, forming by their union those bundles which are seen 
* Philos. Trans. 1853, p. 350. T Proceedings of the Royal Society, Jan. 27, 1857. 
i Histologie Comparée du Syst. Nerv. Paris, 1857. ^ § Philos. Trans. 1853. 
