Several dissections were performed to determine the level of spinal cord termination and the vertebral level at which the dorsal and ventral roots of spinal nerves C1-S4 emerged from the spinal cord in the rat. These levels of emergence were then compared to the level of exit from the vertebral canal. The dissections demonstrated that the effect of differential growth between spinal cord and vertebral column begins in the lower cervical region and becomes progressively more pronounced throughout thoracic and lumbar levels. The disparity between the vertebral level of emergence of spinal roots from the spinal cord and their level of exit via intervertebral foramina was found to be considerably larger than was previously reported by Greene ('68). It was further noted that the spinal cord terminated at the level of the intervertebral disc between the third and fourth lumbar vertebrae, not between the fourth and fifth lumbar vertebrae as reported by Greene ('68).

Transection of the spinal cord is adult Long-Evans Hooded rats is followed by the formation of a connective tissue matrix in the lesion site and the rapid erosion of the neural elements above and below this zone particularly within the dorsal white columns. In the period between 15--45 days after operation two significant events begin to occur. First the injured surfaces of the divided cord become invested by a glial limiting membrane (glia limitans) and, concomitantly, large numbers of axons ensheathed by Schwann cells sprout into the scar matrix and along the eroded dorsal column region. The injured surface of the spinal cord is highly irregular with deep, collagen-filled rifts into which the sprouting axons may probe and penetrate into the adjacent normal neuropil. Electron microscopic examination generally reveals the interposed glia limitans and that these fibres are usually restricted to the peripheral environment. However, as some axons approach the reconstituting glia limitans, they are enveloped by an astrocytic cytoplasmic process which may either displace the Schwann cell or encompass it together with the enclosed axons. This last phenomenon appears to precede the entry of some axons into the neuropil and suggests that the glia limitans may not necessarily represent an impenetrable barrier to the passage of regenerating axons into the CNS. Apparent maintenance of most of these fibres for periods of up to 3 months may suggest that viable, functional synapses were established upon available neuronal elements, but clear evidence of this could not be documented.

Long-Evans hooded rats were cordotomized at the T-5 level and given either (1) cyclophosphamide (cytoxan), an immunosuppressive, (2) piromen, a bacterial polysaccharide-nucleic acid complex, (3) topical and systemic trypsin, or (4) no further specific treatment. Because of past and present controversy surrounding the proposed ability of these agents to promote spinal cord regeneration, a systematic study, employing light and electron microscopy, and quantitative methods in a single animal model, was done in order to re-evaluate the effects of each treatment upon the connective tissue matrix which forms in the defect left by transection. After an initial inflammatory reaction during the first week after surgery, the lesion zone is characterized either by areas of dense collagenous connective tissue with occasional fibroblasts and macrophages, or a loose areolar tissue with numerous sheets and cords of mesodermal cellular elements but minimal collagen. By 45 days postoperatively (dpo), axons supported by Schwann cells invade and become entangled in the loose connective tissue matrix. With longer postoperative survival, cysts appear craniad and caudad to the lesion and erode much of the scar together with viable neural tissue. Giving cytoxan or piromen did not result in any qualitative alteration of the scar matrix as evidenced by electron microscopy. Quantitative analysis revealed a slight reduction in the fibrous connective tissue component of the scar at 45--90 dpo, but this was transient when longer postoperative periods were studied. Trypsin caused a significant reduction in the amount of fibrous connective tissue with a concomitant increase in loose connective tissue and the appearance of a few distinctive, compact bundles of unmyelinated axons lacking Schwann cells. Consistent behavioral changes were not observed in any group which could distinguish them from the controls. Our results appear to contradict the findings of Matinian and Andreasian (1976) who reported return of normal sensori-motor function in 80% of their animals treated with topical and systemic trypsin. It is concluded that a major impediment to whatever longterm regenerative potential exists within the spinal cord is the lack of axonal guiding elements within the scar, but more importantly, the severe erosion of the remaining spinal cord due to cyst enlargement.